The Data Canteen: Episode 10
Molly Larkin: FourthBrain
In this episode, Chris Sanchez and I have a fantastic conversation with Molly Larkin, Director of Operations at FourthBrain. If you haven’t heard of it, FourthBrain is an awesome bootcamp that’s backed by Andrew Ng’s AI Fund. FourthBrain offers hybrid online programs for budding machine learning engineers.
The Data Canteen’s parent organization, Veterans in Data Science and Machine Learning (VDSML), is excited to announce that we have partnered with FourthBrain to support and encourage veterans in pursuit of careers in machine learning. All of our community members are eligible for a $500 discount toward the FourthBrain program of their choice. Additionally, one of our members will win a FourthBrain scholarship worth $3,000!
So, if you’re interested in becoming a machine learning engineer, listen on and make sure to submit your VDSML FourthBrain scholarship application by September 17th 2021.
FEATURED GUEST:
Name: Molly Larkin
Email: info@fourthbrain.ai
LinkedIn: https://www.linkedin.com/in/mollyalarkin/
SUPPORT THE DATA CANTEEN (LIKE PBS, WE'RE LISTENER SUPPORTED!):
Donate: https://vetsindatascience.com/support-join
EPISODE LINKS:
FourthBrain’s Machine Learning Programs: https://www.fourthbrain.ai/curriculum
Apply for VDSML’s FourthBrain Scholarship: https://vetsindatascience.com/vdsml-scholarship
The Batch Newsletter: https://read.deeplearning.ai/the-batch/
Machine Learnings: https://subscribe.machinelearnings.co/
MLOps Community: https://mlops.community/
A Modern, Scalable Approach to Retooling the U.S. Workforce by 2025: https://vetsindatascience.com/articles/a-modern-scalable-approach-to-retooling-the-us-workforce-by-2025
PODCAST INFO:
Host: Ted Hallum
Website: https://vetsindatascience.com/thedatacanteen
Apple Podcasts: https://podcasts.apple.com/us/podcast/the-data-canteen/id1551751086
YouTube: https://www.youtube.com/channel/UCaNx9aLFRy1h9P22hd8ZPyw
Stitcher: https://www.stitcher.com/show/the-data-canteen
CONTACT THE DATA CANTEEN:
Voicemail: https://www.speakpipe.com/datacanteen
VETERANS IN DATA SCIENCE & MACHINE LEARNING:
Website: https://vetsindatascience.com/
Join the Community on LinkedIn: https://www.linkedin.com/groups/8989903/
OUTLINE:
00:00:07 - Introduction
00:02:46 – What is FourthBrain & how did it get started?
00:03:53 – What distinguishes FourthBrain from other data science-related bootcamps?
00:04:22 – How does FourthBrain define the role of machine learning engineer?
00:06:01 – What makes FourthBrain a good fit for veterans?
00:08:26 – Molly Larkin’s story/background
00:13:51 – The expected demand signal for machine learning engineers
00:22:30 - How has the demand signal for machine learning skills evolved over the past five years?
00:25:07 – FourthBrain’s only prerequisite requirement
00:28:53 – FourthBrain’s Python skill assessment
00:36:01 – A summary of the FourthBrain program experience
00:39:27 – FourthBrain’s online hybrid synchronous/asynchronous format
00:41:47 – Future funding & aid for veterans
00:45:12 – FourthBrain’s project-based curriculum
00:50:24 – FourthBrain’s emphasis on technical soft-skills
00:52:24 – How to become the non-existent machine learning total package that employers are looking for
00:55:14 – What types of students naturally excel when going through FourthBrain’s programs
00:58:22 – What types of students tend to struggle when going through FourthBrain’s programs
01:00:58 – FourthBrain outcomes
01:05:49 – MLOps Community
01:14:36 – How is FourthBrain looking to expand its program offerings over the next three years
01:24:32 – Molly’s AI & ML resource recommendations
01:26:14 – Molly’s preferred methods of contact
Transcript
DISCLAIMER: This is a direct, machine-generated transcript of the podcast audio and may not be grammatically correct.
Ted Hallum: [00:00:07] Welcome to another episode of the data, canteen, a podcast focused on the care and feeding of data scientists and machine learning engineers who share the Columbine of us military service. I'm your host alum. And today I'm excited to bring you a special conversation with Molly Larkin and Chris Sanchez.
Molly is the director of operations at fourth brain. And Chris is a fellow member of our community. If you haven't heard about it all. Fourth brain is an awesome bootcamp. This backpack interviewings AI fund fourth-grade offers hybrid online programs for budding machine learning engineers. I first accord this episode a few months ago with the intent of heightening our memberships awareness about this veteran friendly path to mill engineering and offer a meager scholarship opportunity.
However, in the intervening months, fourth brain has worked with us to offer our membership the opportunity to compete for a merit based. Worth $3,000. Fourth brains programs have a tuition cost of $6,000. Meaning that the winners out of pocket tuition costs will be reduced by at least 50%. So if you're interested in becoming a machine learning engineer, listen on and make sure to submit your application by September 17th, 2021 links can be found in the show notes.
Molly. Welcome to the show. And Chris, welcome back. It's so great to have you on again. Since the data canteen launched earlier this year, we've had. Several episodes where the merits of boot of bootcamps were just touted from the mountain tops.
We've had people who went to boot camps had a great experience. In episode five, we had Rob Alberton come on. He's the senior director of Okta's AI center of excellence. And he talked about how some of his best hires have been folks who graduated from data science, machine learning, boot camps.
And then of course, in episode two, Chris, we heard about your personal experience with galvanize and you learned a lot of good practical stuff there that I think you even said that in some of the recent interviews you had, you felt like some of the skills you learned there at galvanize works were crucial.
So having said all that Molly, prior to your current role there at fourth brain, I believe I saw in your background information that you spent two years as director of operations at galvanize, and then another two years as director of operations that you Udacity. So Chris, with you having gone through bootcamp yourself, Molly, you have tremendous experience in this area of non-traditional education between galvanized you Udacity and now fourth brain.
I think we've got a perfect mix of perspectives here to have an amazing conversation on this topic to kick things off. Molly, could you just tell us all about fourth brain for people who haven't heard of it? I'd love to hear how it got started when it got started and what's transpired from then dental now.
Molly Larkin: [00:02:46] Yeah. Thanks. Thanks so much for having me. So forebrain is a program that is trying to get more people into AI and ML. So we were started last year. 2020 right after the pandemic hit is when our CEO started. And so we have one program right now, it's the machine learning engineer program and it is designed to get people in machine learning, engineer careers.
We're part of the AI fund. So injuring or many people are familiar with started at investment studio, all the AI fund. And we are one of the, the portfolio companies in that along with companies like deep learning AI, which many people are also familiar with. So Andrew's not involved in our day to day, but he is on our board.
And so we're very lucky. Yeah. We're very lucky to be affiliated with, with these really amazing other companies. So, so like you said, are we, our first program is the machine running engineer program it's designed for people who have a background in data or programming and then moves them into a machine learning engineer.
Ted Hallum: [00:03:53] Cool. So that's what distinguishes your program from other programs like galvanize, where you were at before is where it as galvanized is more geared towards data science. Generally you guys focus on machine learning, machine learning engineering, right?
Molly Larkin: [00:04:05] Correct. Lots of data science, boot camps out there.
And of course there's overlap between machine learning and data science. But our focus is machine learning engineering specifically, which is relatively unique out there in the bootcamp space.
Ted Hallum: [00:04:19] Now.
Chris Sanchez: [00:04:22] What's that I actually want to ask a question about that. Tamale. How are, how is fourth grade, I guess, defining machine learning engineer earing has a role.
Sorry, but if I stopped,
Ted Hallum: [00:04:33] no, no, I was just about to ask because these terms are oftentimes have conflicting definitions and so that's gonna be super important for this. Yeah, for sure.
Molly Larkin: [00:04:41] That's that's a great question. So I have two sort of different answers to this. So what we think of as a machine learning engineer is somebody who can do the end to end process with the machine learning project. So that means somebody who knows how to collect and wrangle the data who knows how to build and tune a model and who knows how to deploy it and to production level scale.
So we teach that end to end price. Data scientist. I'm definitely going to be working on that data exploration, definitely going to be working on that, that model too Ang probably not involved with the deployment piece. Another way that I've heard described is that a data scientist is always going to include a human in the loop.
A data scientist output is a recommendation a suggestion you know, here's all these different ways that you can do something. And here's what I suggest from knowing that this is use case a machine learning engineer. Their whole point is to remove the human. I'll put a code, not a recommendation.
Ted Hallum: [00:05:45] So it's, it's kinda more of a, at least part of the answer is it comes down to who's the customer. Is the, is the deliverable going to go to a person or is it the, is the PR of what you produce going to be consumed by a machine?
Molly Larkin: [00:05:57] Yeah, that's a great way of.
Ted Hallum: [00:06:01] Cool. Now of course, as you know, the target audience of this particular podcast is veterans.
So I'm definitely interested to learn when it comes to fourth brain, what are the specific things that make fourth brain a compelling non-traditional education option for veterans?
Molly Larkin: [00:06:18] Yeah. So that's a great question.
You know, I think veterans make really great data scientists, machine learning engineers. And we can talk about that a little bit later, you know, and we're talking about racks and the type of people that this program is right for.
But in terms of, you know, why I think this is so compelling to the veteran community or to any community really is are I don't have to tell anybody that machine learning and AI are great fields to be in It's growing. It's, it's a place where some of the skills that, you know, we think are really valuable are going to be coming up with, with this community.
So things like resilience, grit, determination being able to think outside the box, being really methodical in your approach. Th these are all reasons that we think that, that veterans can, can really succeed in this field.
Ted Hallum: [00:07:10] Absolutely. So, you know, I think every veteran is. As they've exited the military service, they want to be in demand.
They want to have skills that are going to make them a hot commodity. And I mentioned Rob Allbritton and Octo earlier, he and I have had conversations and he's told me, you know, that it's not uncommon at all for him to talk to people who've just graduated from graduate school programs. So formal education programs, they've learned all about data science and machine learning in the sense of building a model, but they didn't get any exposure to the deployment of the model, which is such a huge piece.
And you know, to, to the extent that he says, I can't find people who have been trained. Like I want to hire people who have that skill, the ability to take a model and put it in to deployment in a way that's scalable and sustainable, but I can't find people with that skill. So for our veterans who want to make yourself attractive to employers, being able to get this machine learning engineer skill, being able to take a machine learning model.
Put it into production, make it scale and make it sustainable. That's huge. I mean, like, I think you're immediately hireable if you get that skill. So this, this episode is definitely this is gold for, for our listeners who are coming out of the military right now, and they're trying to figure out how to make their way in machine learning.
But before we get too far into all of that Molly, this is your first time on the show. So since you've not been on the ship before, we don't know your background and everybody that's in this field has an amazing, inspiring story to tell. And I don't want to cheat you out of that opportunity. So just real quick, tell us a little bit about where you're from, what your interests are or were as a kid.
So like whether you were in, into stem and math back then, or if that's something you got into later and then kind of just your career progression up to this point,
Molly Larkin: [00:08:55] Yeah. Thanks. So I grew up outside New York city. I went to U Penn for undergrad and I wasn't really like a science person when I was growing up or a math person.
You know, I was pretty good at that in school, but I, I sort of always knew I was going to go into education actually. So my first job out of school, I was a fourth grade teacher in New York city. And so I taught fourth grade for a couple of years. And then I moved out to California and started working for a nonprofit called tech bridge and tech bridge run programs for girls and science and engineering for middle school and high school girls trying to get them interested in science and, you know, introducing them to female role models in the field.
And so that's really what sparked my interest in stem education and, you know, getting more people into these really lucrative and stable. So after that, I, I started working for a very small one of the very first data science boot camps out there called Zipfian academy which was then acquired by galvanize and then came the galvanized data science.
Which is still around today and still a fantastic program. Congratulations, Chris. And so, yeah, so I, I was, you know, doing operations in admissions at galvanize across both the data science bootcamp, the web dev bootcamp. And for a time we were offering actually a master's program in data science as well in partnership with the university of new Haven.
But that program is no longer in existence, but when I was there we were, we were running that program. Then I moved to Udacity just yet another aspect of adults non-traditional education. And I had a variety of roles in admissions, operations marketing as those programs kind of were in flux during those years that I was there.
I joined fourth brain last year the C no fourth brain I worked for at U Udacity. And so, you know, I kind of ended up working with her again and it's, it's been really wonderful. I've been there for about nine months, not fourth grade.
Chris Sanchez: [00:10:54] So, yeah, you mentioned you acid in that, that sparks some memories, because I remember selling glasses is like one of the first, you know, first, one of these alternative education companies to market them. And what was the other one? GA general assembly was the other one that was like kind of big. And I I've taken some U Udacity courses.
I never finished like one of their Nanodegree programs. I was, I was doing my burglary masters at the time, so I didn't feel the need to do both, but I would go over to the gas if you kind of supplement some, some learning, maybe you do some, you know, empire or whatever. But yeah, I just, it, it brought back some memories, but I mean, you've been in that space for five years now.
So, and you were in at the ground floor, right? Zipfian. Yeah. So I'm, I've always, you know, be curious to know about is is what have you seen transpire over that course of time? You know what I guess, like what trends have you seen in that, in that field or in that field of education and yeah. Kinda, how has it changed, you know, over these last five years?
Yeah.
Molly Larkin: [00:12:01] So so I started dipping academy in 2014, so I'm really right near the beginning of sort of the bootcamp. I don't even know the word for it right when we started exploding. And so I was there for like this rise and they were sprouting up everywhere and, you know, everyone wants to take a bootcamp and it was the future of education.
And then, you know, just two years ago, I think is when it kind of started dipping and sort of course-correcting a little bit. But so, you know, I
Chris Sanchez: [00:12:31] think the thing do you mean like attendance.
Molly Larkin: [00:12:38] Yeah, a couple of programs closed. I think in 2018 you know, some consolidation galvanizes, one organize bought hack reactor which was probably the first, you know, well-known boot camp out there.
And so which is normal, that's normal in a new industry, you know, some consolidations you know, reconfiguration I think throughout that time, you know, people were during that time, people were questioning like, are the camps really the future? Are these going to stick around? And I think the answer is a definite yes.
I mean, they're, they're really valuable. They, they offer a valuable service. They, they get people into jobs that they're looking for. And I think, you know, Rob said it in the last episode and, you know, Chris, I think you're an example as well, is that they can work. They're not going to work for everybody, but there is a good population of students for which they work.
So, so I think it's just going to continue being a a format of education especially in tech, as the technology changes really fast as the hiring needs change really fast, traditional education can't always keep up. So, so these really fill a gap.
Chris Sanchez: [00:13:51] Yeah. Our host Ted helm actually wrote a. Wrote an article on the changing workforce, you know, there is some, there was some startling statistics in there.
I think I've got one right here, here. It says 17% of the us workforce is expected to need six to 12 months. Right. So that's, you know, long bootcamp or potentially even like a master's program of skilling to remain competitive for gainful employment starting 20, 25 and beyond. Right. That's, that's a lot of that's a lot of upskilling, right.
And for some people that are kind of stuck in these kind of static roles, like to have to step away from that for a period of time and, and with, you know, the hope that they're going to find something on the other side, there's a good chunk of people who probably won't want to take that leap. But that's a, you know, that's kinda, that's kind of a scary proposition, so.
You know, one of the solutions that Ted wrote in his paper was know the explosion of MOOCs and how useful is that I will be to, to provide upscaling. But I think that this, you know, this bootcamp, you know, this alternative education field is another piece of that puzzle. Right? It's another You know, it's another way of, for people, it's another path for people to get that upscaling that's required.
So, I mean, do you, it sounds like you do think that it's just going to continue to increase in popularity as time goes on.
Molly Larkin: [00:15:25] I think so. And I think I agree with everything you're saying, you know, that number that, you know, this huge percentage of people needs upscaling. It sounds scary, but it doesn't have to be scary.
The, the education is out there. The providers are out there in both muck form and bootcamp form and masters form, you know, it's, it's all out there. And I think everybody is starting to come around to the idea of this lifelong learning. You know, it, it, hasn't been true for a very long time that you get your one degree and you stay at your one company for 40 years and that's it.
You know, that that world is long gone. And I think most people know that and Are are interested in this lifelong learning. We get about 5% doesn't sound like that much, but you know, that represents a few people in our, in our cohorts that say, they're just learning just for fun. They just want to gain skills and they relevant.
You know, they're not, they're not really looking for a new job. They're not even looking for a promotion. They're just interested in learning these things. I think that's important. Those are there,
Ted Hallum: [00:16:27] but those are the best students just
Molly Larkin: [00:16:31] cause they want to. I mean, other people, you know, getting a new job is pretty good motivation as well.
So, so we're, we're really lucky with our students. You know, one thing that's really interesting. And, and Chris, you know, when I was reading some of your notes ahead of time, I was looking at these numbers again. You know, the, the number has fluctuated over the last 30 years, but pretty consistently for the last 10 years, of computing PhDs, Only 30% stay in academia and stay to teach the next crop of PhD. So that means that those number of PhD students has not changed much in 10 years, simply because there's not enough capacity. There's not enough people to teach them. They're all going to get, you know, their million dollar contracts at Google.
Once they get a PhD, you know which is understandable. But I think what it means is that, you know, as AI grow and ML demand grows there's definitely not going to be enough PhDs to fill that gap. So they're going to have to look at other ways of educated people. So, you know, a master's is of course one of them, but these other non-traditional pathways are, are going to be there already are necessary and they're just going to continue to be necessary.
Chris Sanchez: [00:17:42] So where, at what point do you see, you know, again, cause you you've been in the field for the last five years. At what point in the future do you, do you think that we'll start to see job postings for data science and machine learning positions? They're like, Hey, you know, do you have a bachelor's in this?
Or have you attended, you know, maybe one of these top 10 bootcamp. Right. And I asked that, you know, not facetiously, but like, cause I, I I've been, you know, looking for a role for the last three months and hopefully that's going to come to a close next week. But I started to see postings where one prominent organization in particular that I can think of where they said, Hey, do you have, you know, certificates from Coursera?
And I just remember looking at them like, whoa, like that's the first time we've ever seen that. And like, this is like a well-known organization and they're posting that. Like they're going to accept non traditional education. I was like, how cool is that? And another corporation that I actually applied to and move forward in their interview process.
The requirement was PhD or master's, or, or two to four years of experience in, in data science. So you didn't even need a degree, but they were willing to take someone who had just like, you know, had, had hacked together and learned how to code. And I was like, wow, that's some progressive thinking. So where do you, I guess, at what point do you, do you think that we'll start to see, you know, like maybe 50%?
Molly Larkin: [00:19:11] Yeah, that's a really good question. I don't know how much we're going to see it in actual postings. I I've got to imagine it's going to go up. I would imagine that internal processes are certainly changing if they haven't already. I know that there's a few companies that have started to do away with the requirement for bachelor's degree, for some positions, you know, and that's, that's sort of a step in the same direction of, of what we're talking about is this recognition that the traditional signifiers of skills are still great and still valid, you know, like not to knock on the master's programs and the PhDs out there.
You know, those are obviously fantastic programs, but there are other ways to get these skills. And it doesn't mean that that person has, it is not as qualified. It means they are qualified in a different way. And so I definitely think that those internal processes and the sort of internal criteria, even if they're not necessarily reflected on a job description are changing.
Chris Sanchez: [00:20:11] As one real quick as an aside and Ted, I know you'll get a kick out of this one organization that I know that isn't going to pick up on that trend. Fast enough. I don't think U S government, right? So I applied through a DOD organization and move forward in their interview process and they knew that I had a master's and they actually, you know, wanted to give you an offer, but they said, Hey, it's contingent on.
Their HR department being able to take a look through my coursework and see that I had met the requirements for like calculus and linear algebra. And I was like, so masters of data science, look at my good portfolio. It's pretty, it's pretty clear that I know, you know, that I meet the requirements, but they still, it didn't matter.
They had, I had to like check these certain blocks and, you know, that was at that one organization. So I gotta imagine that that's, you know, across DOD. And, but that kind of thinking, you know, is that's, that's 1995, right? It's we're, you know, we're entering the age.
Molly Larkin: [00:21:17] Yeah. There's always going to be industries and there's always going to be companies that.
Stick to tradition and, and they want their, you know, their hires to have the degree from the top 10 school. We're always going to exist. But there's a lot more companies that don't have the requirements, you know, from your perspective. And, you know, obviously. The veteran community going to work in government or government related fields is probably higher than the normal population.
So that is certainly a concern, but you know, hopefully that, you know, just like all the other industries are sort of shifting this direction at some point government will to maybe
Chris Sanchez: [00:22:05] shift fast enough.
Ted Hallum: [00:22:06] I totally agree. What's to say, I think they'll, I think they'll get there. It might be a decade and a half from now after so many good people I've gone to work elsewhere when they could have been put to work contributing to our national security. But, you know, I wish I had more control in that, but it's not something that I have any control over at all.
I can just hope for the.
Chris Sanchez: [00:22:30] Yeah. All right. Last question on this topic, Molly is what kind of skills have you seen a wax and wane over that period of time? You know, from 2015 til now, from when you kind of started at galvanized, like kind of what were the skills, those job market we're looking for versus what are they, what are you seeing now?
Molly Larkin: [00:22:52] Yeah, that's a great question. And I think it's, it's less than the skills have changed. It's more that the definitions have changed. So. 30 years ago, 25 years ago, we had webmasters they did UX, they did UI, they did content management. They did the programming, they did the HTML. They did everything related to your website, right?
There was no really differentiation of, of what that person would be doing. Now, obviously all those things are different roles and they're doing specializations and there's different tools and skill sets to fill all of those roles. And I think that's sort of where data science and machine learning is going.
You know, eight years ago data science was the hot new job title, right. But it could have covered everything from analysts to statisticians, to machine learning, researchers, you know, it covered this huge range of roles that weren't really defined necessarily. Well, I don't think we're, we're, we're getting better at that we're data scientists still ill-defined role machine learning can still be an L defined role at certain places.
But I think what we're getting more into is, okay, this is what a data scientist does. This is what a machine learning researchers does. This is what a data engineer does, and just sort of differentiating those roles in those skill sets a little bit more clearly, I think is the biggest trend that we've seen over the last five years or so.
Ted Hallum: [00:24:21] Boom mind blown. Like that was a perfect analogy. I totally get that. I was actually like tinkering with web design back when webmasters were called web masters. I know that I just dated myself, but that's totally true. Like I had to do all that stuff, but none of those other more specialized titles existed.
And I definitely see what we started off calling data science, going down that same kind of evolutionary path. So that's awesome. That was a fantastic perspective. I'm so glad you asked that question, Chris. So Molly getting into fourth brains program for aspiring machine learning engineers for folks who are listening, I just kind of want to go through this in the same way that a prospective student would probably approach the program.
So the first thing that a prospective student is gonna want to know about is the prerequisites, because that pretty much defines whether they're in the running to attend the program, or if for whatever reason they wouldn't be eligible to pursue it. So for students that are interested in getting into fourth brains machine learning engineer program, what prerequisites are required and are they rigid or flex.
Molly Larkin: [00:25:32] That's a great question.
So our only non-negotiable prereq is Python for our program, as it is right now, our machine learning engineer program, you have to know Python, we're not teaching it right now. We do expect you to come in at a particular level. And it's something that will be assessed during the admissions process.
So that is something that, that you, you have to have, you can't learn it during the program. Outside of that, not too many strict requirements. Most of the students that come to us roughly 70% are coming from a data or a software background. So they're either software engineers or data scientists or data analysts, and they're sort of making that progression into machine learning engineering.
Those are coming from outside of tech. We've had musicians, we had high school teachers come through our program. We've had academics finance, quants, you know, really a wide variety of people. And we really, really value those people who are coming from outside of tech. I can't emphasize this enough.
This idea that, you know, you have this really, you know, specific domain knowledge in a, in an area that maybe hasn't been, you know, infiltrated by ML yet. And you're looking to kind of add ML skillset to your existing skillset is really valuable. So we, we think that these perspectives are, are really important to bring into now.
And that goes for people with, you know, a military background. This is like really specific domain knowledge is so crucial. You know, when you're, you're coming either into ML and working for a tech company or possibly bringing ML back into where you, where you want to go. That that's really crucial for us.
One other thing too, I'll just sort of emphasize our, our founding team is a hundred percent female, and we really have a focus on underrepresented groups. We, we really want, you know, people who are underrepresented, underrepresented in tech to come to us to come and apply because we don't have specific professional background requirements.
We don't have specific educational requirements. It's just Python. So, so come check us out. We we've partnered with groups like AI inclusive to offer scholarships. So. Something that's really important to us as a, as a team.
Ted Hallum: [00:27:56] Well, before we dive into curriculum, you know, since you mentioned that when I think about veterans, a lot of times, you know, the defining characteristics that are most common among veterans that are separate from the service, it's, it's unusual to find a veteran that is not good at taking initiative.
And it's unusual to find a veteran that's not resilient. So from what you just said, my understanding is that if there's a prior service infantry, Ben artillerymen sailor, whatever the case may be, and they are willing to take the initiative and go out and, and fiercely and resiliently pour themselves into Python, which may be completely alien to them.
And they get a good working knowledge of Python and they can put Python to use. Then your program is wide open to them. Is that.
Molly Larkin: [00:28:46] Absolutely. We have we've had people come through the program who are self-taught Python programmers, and that's just fine.
Ted Hallum: [00:28:53] Fantastic. Now the last question on that I did hear, you mentioned a, a Python assessment to, to sort of gauge the amount to which someone has learned Python.
Can you speak to what that assessment process looks like? I'm sure some of our listeners would be curious to know. Sure.
Molly Larkin: [00:29:10] It's pretty basic. It's on, we use a platform called coder bite which is similar to hacker rank or day request. It's an algorithm challenge. So, you know, it's sort of general purpose Python programming.
One thing to know is we're a school, we're not a job. We're not looking for a hundred percent. And if, if you don't, you know, do so well the first time you just try it again, it's fine. Or if you say, Hey, you know what? I think I need to take another month and keep working on some of these skills and come back in a month.
That's great too. This is not like a one and you're done type of situation. It's, it's more, we want to make sure that you are set up for success in the program and that when you come to us, you are ready to learn machine learning. You're not still struggling with Python libraries.
Chris Sanchez: [00:29:59] I, Jen, I've taken it cause I actually applied to fourth grade last year as part of my assessment of looking at different, you know, boot camps.
And it's tough. Like you like you couldn't, it wasn't like a, well, let me sit down and bang it out in 15 minutes. Like, no, I remember it took me three, four hours to solve it. And a lot of, you know, a lot of trial and error to get it done, but you know, to your point, Molly you had to know Python, you know, you couldn't, if you were looking at syntax stuff like you, weren't gonna be able to finish that.
And I think that that's a really. Smart thing to do, because in terms of, you know, focusing as that on your assessment, because in my experience, it's the ability to express your thoughts into code. It really opens up the fields, right. Because if you're having to learn like two things at the same time, like the actual concept, and then trying to apply that concept of code, like that's, that's not fun.
I've been there. It's not a good, you know, it's not a good way to learn.
Ted Hallum: [00:30:56] So I gathered you have to solve a challenge. Is it timed?
Molly Larkin: [00:31:01] No. Okay. Okay.
Ted Hallum: [00:31:04] Nope, not timed. Cool. Yeah.
Chris Sanchez: [00:31:08] Okay. Yeah, so, you know, Molly, I'm glad that you cleared that up because that's all I actually was on the phone the other day through through with a veteran on the Vetter ADI platform.
He, he called me up and I actually recommended fourth brain as the path for him. Yeah. Cause he's a, he's got a software development background. But he wants to get into machine learning. Right. Engineering specifically. He said, those were specifically, I was like, well, I like there's this one program recommended.
I was like, exactly sure. What the background that was required. Right. So I, that, that was the part that I was kind of confused about was like, do you need to have like this strong data science background to be successful in your program? And it sounds like you don't
Molly Larkin: [00:31:53] You don't, but most students do, you know that they're following a really relatively traditional path.
But it's not required. And like I said, some of the people who've come to us who are coming from outside of tech who are self-taught programmers, they're quite exceptional. You know, they they've taught themselves Python on their own. They've been learning these things on their own. And so they're, they do just fine in the program.
Just like anybody it's like somebody with a traditional background. So
Chris Sanchez: [00:32:22] I it, I that's. So as we're talking about it, it's really interesting to hear fourth brains take on machine learning, because I see that at Google too, where they're just like, Hey, machine learning is this technology, right? And as long as you know how to code, then you can use that technology to do X or Y in this particular field.
And it sounds like that's worth grades approach, as opposed to like the kind of, I guess the way I was brought up, maybe where it's like, oh, machine learning would be a set of statistics and data science. So if you don't have like a data science background, you don't know what you're doing. Kind of like, you're kind of just say, Hey look, the technology has evolved to the point where you can as long as you don't have to program, you know, you guys understand some basic statistics that you can actually apply it, you know, it's applied ML, right?
Molly Larkin: [00:33:07] Yes. And I think both the technology has evolved and I also have to say the mindset has evolved. So I think that Historically fields like computer science or machine learning have been very gatekeepery, you know, there's a set of people who have the skills and it's, oh, it, this is very difficult. You need to get it.
Like, you need to learn all of these things to come and do this. And I think that that mindset is evolving, not just in tech, but in a lot of things like, Hey, no, this is not that difficult. It is difficult. You know, you do have to study, you need to learn a lot of skills, but lots of people can do this if they're given the chance and the opportunity and, and it's not locked behind 10 years of higher education.
Ted Hallum: [00:33:54] Yeah, no, I think that's the culture that can exist when you're in an AI winter, you know, when the skills aren't really in demand, you can have your little elitist, exclusive club that nobody can be a part of, but when the entire world needs this technology and they need it applied then that the paradigm kind of gets turned on its head.
Molly Larkin: [00:34:15] Yeah. You know, all of the, oh, go ahead, go ahead. I was just going to say, you know, all that being said, you know, clearly there are some roles that are suited for people who do have 10 years of, of research experience and that's fine, you know, that there's nothing wrong with that. And I'm not saying that that's, that is not true.
There are those roles, but there's a lot of roles that don't require that. And, and those roles are growing.
Chris Sanchez: [00:34:44] Yeah. Now I, yeah, I think, you know, there's definitely a spectrum, right? So from the machine and a researcher, who's going to develop, you know, an improvement on Adam optimizer all the way to the, on the strictly applied AI guy.
I know how to, I don't know how to build a microwave, but I know how to turn one on. And that's all I really need to know for this particular use case, you know, in my particular field. So let's dig into the curriculum a little bit machine learning, hearing, right? I'd imagine it's you gotta have, you gotta have knowledge of machine learning algorithms, some software engineering.
So machine learning operations, right. So getting more into the production side of it and then soft skills communication. Like, let's say you've got, you know, Joe programmer or Jane program. They've, they've been coding in Python for a year. They know how to write a you know, how to write a blackjack program or something, but they don't know how to calculate a variance.
Right. So what kind of in the curriculum, what kind of, you know, I guess data science background or, or. Statistics and probability background, do you guys get into, and then how do you build on that throughout the course to the point where, you know, where the students actually deploying right. A project is their final thing.
Molly Larkin: [00:36:01] So so we're very focused on applied AI and practical AI. So we are, we're not necessarily diving deep into the math behind things. So we don't require a math degree or a statistics degree. Obviously having those skills is very useful. And, and you'll understand a little bit more of what's going on under the hood, but what we're really focused on is like, Hey, look, these models exist.
You can use them. Here's how to use them. Here's the use case. Here's how you tune them. Here's how you make them more for your data. So it's very much an apply perspective. For the first half of the course. The second half of the course is focused on the ops and the deployment piece. So I'm in the second half of the program, you do a capstone project and to end where you kind of, I've been to one of the, of the use cases, you pick your own topic, we help you scope it.
And your final product has to be a deployed solution. And AWS that works on, you know, large scale data that that's accessible. And that deployment might look different based on what the, what the project is. And it might be as simple as an API. It might be as complicated as a full front end, you know application.
But the, the point is that you can do the full process end to end.
Chris Sanchez: [00:37:25] Okay. And I guess, like in terms of the, you know, the foundation of all, everything that we're talking about is obviously statistics, you know, these algorithms. Yeah. Are some of them are written like in the fifties, right. They just would just send up, didn't have the compute power to actually make, you know, make use of them in a practical application.
Is there, is there a kind of, some kind of fundamental groundwork that, that your curriculum.
Molly Larkin: [00:37:48] Yeah. So, so we do cover the statistics behind the models enough so that you understand what's going on. We also the pre-course curriculum. So before you even come on day one, we have a little bit of practice in probability and statistics, calculus, and linear algebra kind of just a refresher, get you on the same page as everybody else.
It's not as extensive as a full course in those topics, but it's meant to get you, you know, no one, not to, to be able to apply the models without doing it blindly.
Chris Sanchez: [00:38:19] Okay. Yeah. It's not, you know, as we're talking about this, it reminds me a lot of fast AI, which I know you're familiar with where they're saying, Hey, they kind of have that same philosophy.
Like, Hey, we're going to teach you from the top down. This is how the model works. This is how you make it work. Now, go forth and go into your respective fields and apply it on. No x-ray images or go apply it on your, you know, your natural resourcing, you know, images that you're seeing. And then once you've done that, then come back and then we'll go underneath the hood and we'll kind of take a look at what's actually happening.
But yeah, there's been some pretty cool stories from that. You know, people have actually gone forth and created applications that, you know, may, you know, move the needle in their respective
Molly Larkin: [00:39:00] fields. Yeah. W w what we basically envision is somebody at a smaller mid company. They start as a machine learning engineer and they, they just need to make it work.
They need to figure it out and needs to be good enough. You know, we're not necessarily training machine learning researchers. That's a different skill set and a different focus area where we're focused on the engineering, somebody who can do and build
Chris Sanchez: [00:39:27] nice. And I forgot to ask this earlier, but like, in terms of logistics, is this a, from other call correctly, this was a Saturday based course, right?
So. Yeah, I guess you could consider this a part-time program in the sense that you can still work full time while you're going through fourth grade or
Molly Larkin: [00:39:45] correct. It's it's a part-time program. Nearly all of our students have full-time commitments, either work or school. It will be a busy four months for you.
I'm not gonna lie. We expect that you'll spend about 15 to 20 hours a week on our course. But yes, six hours of that time is a live class session right now. We offer them on Saturdays in the future. We will likely offer them in evenings. So, but outside of traditional work hours and that there is that, that live class component.
We think that this model is, is really effective, just, you know, from all of our staff has experienced in non-traditional education. And so the sort of combination of the flexibility of online combined with the accountability of an instructor and a cohort and the camaraderie of going through the class together with a cohort of your peers, all of those pieces, we think combined to make this a really effective learning experiences.
The model that we thought was was the right one for fourth brain.
Chris Sanchez: [00:40:48] Nice. And I remember, so I remember when I applied, like you said, this kid right during the pandemic, and I think originally. You had planned on it being strictly a physical program, like you had to be in the class. Right. And then COVID said, otherwise is how you're going to continue doing that.
Is it going to continue to have near remote program? Are you going to have like a hybrid option where you can go in person, if you want through what's the, what's the future, hold in that regard,
Molly Larkin: [00:41:16] our immediate plans are to keep it remote. One of the, the really nice benefits that we've had people across the country and across the world come to our program.
So people from south America, we have somebody in India, somebody in Kenya, and of course right now and then just across the U S most of our students are, are US-based. But you know, they can be in California or they can be in Pennsylvania and it's fine. And everybody's in the same class, which is really nice.
Chris Sanchez: [00:41:47] Nice. 10, I don't know if you had covered it earlier, but did you want to talk about funding. For the veterans
Ted Hallum: [00:41:58] We didn't actually, but absolutely I should have asked that. So for your program, if people have GI bill benefits or vet tech benefits or whatever, is your program eligible to apply those veterans education benefits?
Molly Larkin: [00:42:13] Not yet, not yet. We are planning to apply for, for all those programs, as soon as we're eligible. Many of them require you to be in operation for two years before the institution can apply. But that some of those rules have changed still sort of in flux, but we are monitoring all of those and we, we will be applying for inclusion in those, those programs as soon as we're eligible.
Ted Hallum: [00:42:40] Okay. Well, so often, you know, these podcast episodes there. In the ether. And so I can imagine veterans coming across this episode two, three years from now, what's the projected date. When do you guys think you'll be able to accept those veteran education funds?
Molly Larkin: [00:42:56] Yeah, so we launched our operations in October, 2020.
So likely at some point in 2022 is, is my best guess at this point in March of 2021.
Ted Hallum: [00:43:13] So in the meantime, while our listeners probably can't use their veteran education, but if it's, what is the total cost and are there other Financial aid options, scholarships that you know of that could help people with the cost.
Molly Larkin: [00:43:31] Yeah. So the cost of our program is $6,000. And we, we do have payment plans, so we can spread that payment out over a longer period of time. We usually have at least one or two scholarships for each cohort that are done through that are done in partnership with other organizations or sometimes on our own the best thing to do.
Apply. And, and then, you know, reach out at this point in, in our last cycle, I am the person that prospective students talk to. So I'm always happy to talk to you about, you know, your situation, what can work for you. We're, we're very open to kind of coming up with some solutions that work and, and as we grow and kind of continue to evolve, we are hoping to have more formal scholarship and loan programs.
W we aren't an accredited program, so we don't qualify for some of those traditional low options. But there, there are organizations out there that do loans for, for bootcamps and things like that, that we we'll we'll at some point be.
Ted Hallum: [00:44:38] Awesome. Thank
Chris Sanchez: [00:44:39] you. Yeah, they're going to be, I think that the fourth brand program is gonna be a slam dunk for the vet tech program, which as, you know, just plussed up another 20 million.
And I'm not sure if that starts in fiscal year 20, 21 or for if they're pushing that to 20, 22, but that's going to be a huge, you know, once you, of course you guys, once you guys get, you know, on their verification
Molly Larkin: [00:45:03] list or whatever yeah. We have to get verified. Yeah. So we are, we're, we're monitoring that as soon as we can, can apply.
Chris Sanchez: [00:45:12] Yeah. Yeah. Okay. Very good. So switching gears a little bit So Kristen gets in, there you go through the course, they, they have this great, you know, project that they're deploying. Are they one of the, I guess, one of the awesome benefits of going to galvanize which I didn't fully appreciate actually until I got him through the interview cycle was just the, how many?
So we did three projects in galvanize three, like, you know, full scope. You know, I was the only person working on projects, but in between we did these things called case studies. We do them like every other week and it was with a team of people and it would be like a 24 to 48 hour sprint, you know, working on some, you know, a particular use case.
How is fourth grade setup? Is it one, just one big project at the end? Or do you have several projects intersperse throughout? And the reason I'm asking is because I'm going to get into the value of network portfolio.
Molly Larkin: [00:46:14] Yes, absolutely. So, yeah, so all of our curriculum is project-based. So I've, I've alluded to sort of part one part two in our program, right before part one is the what we call the individual portion.
So everybody's doing roughly the same thing and they're generally working on their own. So there are four projects that happened in part one they're spread out over. So some of the projects are spread out over a few weeks and the bulk of the bulk of what you do, actually, the work that you do on that project actually happens during those Saturday sessions.
So you're actually in breakout room, you can work with other people, you can share screen, you can help each other work through these, these mini projects, but ultimately you submit your own. And so for example the first four weeks are all related to e-commerce examples and there are actually four parts of a larger project.
So over the course of four Saturdays, you'll do four components of something that ends up being a larger project that you would, you could put on that hub as, as a project. And there are four of those in part two, the capstone project is really the meat of, of what you're, you'll be demonstrating. And you can actually see our graduates final projects on her blog.
So you can actually go to fourth brain's blog and you can see our first cohort and what they did in their get hub reposts. So we completely agree with you, Chris, that, that get hub repo is a really crucial piece in your progression and sort of your self advertisement of yourself as a machine learning engineer.
So we put a lot of emphasis on that for our students and how to build it, how to write a, read me what it should include. And sort of what that, what that final project repos should look like.
Chris Sanchez: [00:48:02] And is that, is that I guess kind of baked into the curriculum. So there's like a separate thing. Like, Hey, this is how you do it, read me, and this is how you should post your projects.
Molly Larkin: [00:48:14] It's sort of all of the above. So there are some things that are sort of in, in written documentation and we might send you some examples of things that you can look at. Some of it happened during class, and then we've also had some optional workshops. Some of our students who are coming, who are, have a really extensive software background, they, they know how to get hub works.
They know what a repo looks like. So they're, they're familiar with it. Some students who are new to it we actually offered it's separate workshop and like, here's how you use get hub. Here's how you should set up your repo. Here's how you set up your account. And try to emphasize that. So try to meet all the students where they are what their background with things like LinkedIn and get hub NATO, the U S.
Chris Sanchez: [00:48:55] Gotcha. Yeah. For our listeners out there that are either looking to get into the, into the role now or in the future. I found that having a nice blend right of projects, you know, on your portfolio deal, that's your experience, you know, things that you've actually done. And then I guess kind of the book knowledge to pass an interview, you gotta, you kinda gotta have both, right.
Because when I got out of galvanize, I actually didn't focus at all on the, on the book learning part. Right. I just was like, oh, I just got all these projects and I'm just gonna continue making more. And then I had I did my first couple of interviews and one of them, you know, one of them being a mock interview and they were asking some detailed song.
I was like, ah, man, I don't know. Perfect. Well, I can tell you all about the project, you know, all about the data. That's where that I just built, you know, so, but on the other hand, When you get into an interview and they're asking you about your experience, do you know, using this tool or your experience, like if you haven't done that, you can't, you can't fake that, you know, so yes, you may know how dropout works and you may know how to code up you and Adam optimizer.
But if you don't have, if you haven't actually built a deep learning neural net and applied it to something, you can't really talk about that experience. So to listeners, you definitely want both. I can't say, you know, what percentage is over the other. But don't do what I did, which was neglect of one over the yeah.
Molly Larkin: [00:50:24] One of the other things that we explicitly include in our program is practicing communication. And Chris, you, you alluded to this before, and I didn't quite get to it in my answer, but I just wanted to kind of circle back soft skills and communication are really important.
And that includes both verbal and written communication. It includes both formal and informal and includes to both technical and non-technical audiences. So the second half of the program where students are working on their project, we actually have like explicit practice than that in terms of you'll be doing a standup with your small group every weekend, you'll be creating a technical written report.
You'll be expected to give a midpoint presentation and a final presentation and sort of just incorporating all these different ways to be able to succinctly explain what it is you've done and why you've done it. And practicing that lots of different ways. And lots of different times is, is a really crucial part of being successful nuts in interviewing, but in the job.
And so we, we, we incorporate that practice as well.
Chris Sanchez: [00:51:31] That's huge. And I'm glad to hear that that's part of the fourth grade curriculum, because you know, it was, you know, being able to break down yeah. The technical concepts to a non-technical audience. That's kind of, that's really, that's the other half of data science.
Right. And if you spent all your, you know, you spend all your time in a Jupiter notebook without actually having to explain your results to a decision maker near, you know, your, your infrared code, if you've ever had to practice that. One of the mock interview questions that came up was explaining how momentum works, you know, to a non technical person.
And that was, Ooh, how would, how would I do that? So that's good that it's here. That's like, it's kind of interspersed throughout the curriculum, so it's not like, oh, you're here. Here's, here's your one class on it, you know? And they're like, no, you get the chance to actually practice it
Ted Hallum: [00:52:20] throughout. Yep.
So Molly, as I heard you answering some of Chris's questions about the curriculum and about the prerequisites, you know, I get that you're. Most typical student is somebody who has learned Python. And then you're looking to take them down that road to become a functional, good enough machine learning engineers.
So they can, like you say, go into a smaller mid-sized company and be successful. And I think that's fantastic, but also as my mind keeps going back to some of the things that Al Rob Albert, and said in my interview with him he said that he has so many people that he, that applied that he interviews and they've gone through graduate school programs, and they have this thorough grounding in statistics and linear algebra and calculus and machine learning models.
But they've never done the deployment piece. So while I think this is a fantastic learning opportunity for those people who are just, you know, like they're coming in with just that Python foundation, and this is going to get them. You know, that, that base level to be a machine learning engineer. I think that this is an incredible opportunity for like a follow on like the crown Juul and the up-skilling crown of somebody who's just finished a master's degree.
And they got all that background, which is going to be incredibly informative for the stuff they learned in your program to go ahead and do fourth brains program and then learn the deployment piece, which is what Rob said, people don't, they're not getting how to take a model and deploy it production and maintain it.
So it strikes me that, unfortunately right now, if you want, like to be the, the, the, the full meal deal and have everything from, you know, the core foundations mathematically all the way through being able to. Deploy and maintain models in production. That's really what you need. You probably need to do a graduate program and then just immediately cascade straight into fourth brains program.
And then you're going to be like the non-existent total package that every employer is looking for.
Molly Larkin: [00:54:27] Yeah. So, so we have students in our program who are coming off of masters in analytics or masters in data science. So they are looking for that more practical application. Of ML. So the deployment piece is somewhat unusual.
There are other programs that do it. I'm not claiming that we're the only people who teach deployment. But it is more unusual. And so, and it is something that we see was a gap and something that hiring managers were really looking for. If we are lucky enough to have really great curriculum partners and advisors who are willing to give us back.
And so th that was one of the biggest things we heard. They, we, they need people who can take that, that tuned model and make it accessible, make it work for the company and for their teammates.
Ted Hallum: [00:55:14] To shift gears a little bit. I want to talk about students as they come into your program because they're people and as human beings, they have unique personality traits, unique learning styles.
So I know between, you know, obviously you've got experienced in your current role at fourth brain, but you have all those other relevant experiences that you Udacity and galvanize. You've seen many students come through and you've seen the ones that Excel and they start a program and they just knock it out of the park and, and they S they sprint like they were made for it.
And then you've probably seen some other people who, who, who struggled a little bit more, maybe it wasn't the best fit for those people that just really went into the program and killed it and just did awesome. What are some of the unique traits? Predispositions or experiences that those people tend to have.
Just as kind of like, I, I want to have like a yardstick for our listeners. So as they measure themselves, they kind of know are they a good fit? And then, you know, also people can course correct. So maybe if they, if they hear what you have to say and they think, oh, you know, I'm not, so I a good fit then, you know, maybe there's some things they can do to bring themselves more in line with the right template of successful student.
Molly Larkin: [00:56:24] Yeah, absolutely. And some of these things you, you sort of mentioned before, but some of the things that, that really indicate success in a bootcamp student or a non-traditional education student, or somebody who has a lot of grit and persistence in a bootcamp, things are not necessarily handed to you.
You, part of the point is having you figure it out which has meant to, to mimic the real world. So it, it's not a, it's not a bug, it's a feature of the program. The ability to be versatile and Pivot in your approach. So if you're going down a path and you realize it's not working, can you backtrack, can you try something else?
Can you sort of reorient yourself towards the problem and try it in a different way and be willing to do that and be happy to do that. That is part of the learning process. And so if you're looking for like a straight and narrow path the bootcamp style, maybe isn't for you or if that's not where you're going to encounter most of the time and last being organized, being being methodical.
And that's really more of a, not necessarily a bootcamp thing, but just this field. Being able to backtrack, to find your mistake, be methodical in your approach. As you're trying new things, I think all of those things lend themselves well to somebody who's successful. I think that differentiates it from like a master's student.
This can be the same person at different points in their life. I also want to emphasize that, you know, this is not like, oh, this type of person is right for a bootcamp. It's if this is you in your like mindset professional experience right now than a bootcamp would be good for you right now, but that maybe in five years, it's something different and that's okay.
So if you're willing to put in a little bit of workout time and be willing to learn on the job afterwards boot bootcamp is, is a really great choice.
Ted Hallum: [00:58:22] Awesome. Now kind of the next question is the flip side of that. So as you've watched students, are there any recurring commonalities between students or are maybe a less optimal fit, they come into the program and you just watch them struggle and struggle and struggle and, and you know, maybe some of them don't make it.
Maybe some of them make it and just barely make it. Is there any pattern in those students?
Molly Larkin: [00:58:45] Yeah, I would say students who are looking for a lot of direction and handholding in both the prep work and, and where they're going to later. And that's not to say that that's a bad thing. It might just be where you are in your career.
Right. Then like, yes, I know I need, you know, a little bit more structure in, in this whole span and this whole process and that's great. That's why masters programs exist. They're, they're really fantastic at that. A bootcamp. You're going to have to do a little bit, work more work at the beginning. And you're going to have to do a little bit more work on your, your, your first job.
So if you are a student who really just wants a black and white answer, who just says, I just want to know what the right thing to do is, and then I go do it, the bootcamp isn't going to be for you. Same with the student who just wants to kind of check off their list of tasks and then get that piece of paper at the end.
If you're just in this, for that certificate, that's the point of this? Boot bootcamp is the epitome of a program in which you will get out of it, what you put into it. And if you put a lot into it, you will get really amazing returns.
Ted Hallum: [00:59:57] I think that's incredible advice. And I especially liked that last one because I've actually found that to be true out of virtually every experience in life that you are going to get out of it, what you put into it.
So isn't that, it's not a surprise to me that that's the case with your bootcamp. I can see where that would be doubly or triply true in a bootcamp. But I think people should think that about every learning opportunity and data science, machine learning. If they come to whether it's my mind goes to something like data camp, you know, you could blow through a data.
Course and not learn a thing, or you could go through it and learn a whole lot. It, you know, it depends on the mindset you have. Are you just trying to plug in, you know, are, are you letting your mind go on autopilot? And you're just trying to plug in the code that you think data camp wants. So to let you proceed to the next screen, are you actually thinking through what it is that they're trying to teach you?
So
Chris Sanchez: [01:00:50] the data quest, you can just blow through their modules. You actually
Ted Hallum: [01:00:58] sure. So the next part, I think is super exciting because we're going to talk about outcomes and that's what everybody wants to know. That's, you know, while everybody wants to learn or they should want to learn, like we just talked about as they go through the program what they really are, you know, th the, the, the longterm goal is being able to be gainfully employed, being able to be employed in a position that they find fulfilling and satisfying, and that they can be proud that they do and tell their kids, you know, Hey mom, dad does this cool thing.
So With every program, whether it's a bootcamp or a formal education program, they all have a different gravitational pull. When it comes to where their students tend to work geographically, the industries they tend to go to, or sometimes even specific employers, you know, there are some schools and programs that giant consultancies recruit from heavily.
And, and, you know, I've talked to some people who found, you know, to them, they felt like they needed a dimension that because those, those companies hired so many of their students. So I, I would like to know in terms of geography, first of all, where, and I'm sure you, you probably have students all over, but the top three areas geographically, where do students tend to end up after they finish the program?
Therefore. So
Molly Larkin: [01:02:16] I can tell you a little bit more about our current students. Our very first group of students just graduated on month ago. So we don't have any official outcome statistics yet. Although we are very eager to share those, as students are starting to land jobs, we have a lot of students interviewing right now, and we're really excited where they're going to end up most of our staff.
So I, what I can say is that from our first group group of students three students got new jobs during the program. One got a promotion during the program, which is what they were going for that's before they even finished before they finished. Yep. Just as part of their, their job search. And, but then we have lots of students right now we're in the interview process.
And so, like I said, we're recording this in March, 2021. Our first group graduated in February, 2021. So I'm still a little early to give definitive answers to some of those things. But geographically, our students are located across the country. They are heavily concentrated on the east and west coast.
But you know, they're, they're across, across the country. And also really interested in a variety of fields. I mean, obviously many of those ML roles out there that are students along with other, you know, aspiring ML engineers, they are going to be in tech and that's fine. But we're, we're of course seeing, you know, the, the growth of the melon, other industries and other people can speak to this more than I can, but, you know, healthcare manufacturing are two big ones that come up education a little bit more now, too.
So seeing ML percolate out through our industries, I think is where a lot of our graduates are interested in building. One of the things that I think is interesting to think about when you're thinking about a bootcamp versus master's program, is that when you come to a bootcamp. You are the bootcamp plus everything that came before coming to the bootcamp and a race, your prior history doesn't erase your prior experience.
Neither does a master's of course, but when you go through master's program, often the mindset is I'm jumping off from this masters. And I lead with I am and that I have a, an S in computer science rather than S and data analytics. When you come to a bootcamp, you're, I'm a project manager and an ML engineer.
I'm a teacher and an ML engineer. I am a software engineer and an ML engineer. You are both at the same time. And so that really opens up a good mindset about the types of roles and the types of industries and the types of companies you might want to go into.
Ted Hallum: [01:04:53] So it kind of allows you to just like put an exponent on whatever momentum you already had, whatever career momentum you brought into the program.
Molly Larkin: [01:05:01] Of course. And I think that's another unique thing about a bootcamp is that we do have students who are brand new to tech. They're not necessarily going after the same job as the person coming in was 10 years of software experience. Should they be going after the same job necessarily? And that's okay.
You know where we are adding this skillset, we are helping you add the skill set to your resume. So if you're a software engineer for 10 years, and now you have the ML skills, that's just putting you along this, this trajectory that, that you've already started yourself. If you were fully pivoting into this career, maybe your first job is going to be a junior ML position and that's, that's fine.
You know, that, that's great that that's your first job in ML and, and we are happy to kind of get you on that track.
Ted Hallum: [01:05:49] Now I'm going to have to take us down a quick rabbit troll real quick, because you just cued it up too perfectly. So are you familiar with Emma lops.community?
Molly Larkin: [01:05:59] I'm not,
Ted Hallum: [01:06:00] no. Okay. You're, you're going to want to be, so I discovered them about a month ago.
It is a whole community. You can get to the actual domain. Like the website website is dot community. Okay. They have a website, they have a podcast, they have a slack workspace. I believe it's based out of the UK, but it's all about ML ops. It's all about people that their current role, their current responsibility is building machine learning, operations pipelines.
And I was listening to one of the, I don't think it was the most recent, it was within the last three or four episodes of their podcast. And that whole episode of the podcast was that the topic was the lack of, or the very limited number of program and project managers in MLS. And how while there, you know, the a there's very few of them and B w when you compare it to fields like software engineering, there have been people that are held up as like the paragons of what it means to be a program manager in software engineering.
And they said, you know, There hasn't yet been that for ML ops, you know, there have been people kind of limping along and figuring out how to be a program manager or project manager for ML ops, but there's nobody that's just really been held up and said, this is the template for how you do it. Right. So I think that, I think that's really interesting.
It sounds like that that there's just nothing but opportunity if you get the right skills, which they could get through your program. And then there's this huge demand signal and no supply. So that's how they can see that that's how they can become the supply for that demand.
Molly Larkin: [01:07:43] Absolutely. I agree with everything you're saying, and we're also seeing the signals that that ML ops is something that's in really high demand and it's, it's actually, you know, right now we, our program is focused on machine learning engineering, and it is appropriate for a lot of people depending on where they want to go.
But we are planning to launch additional programs you know, this year, next year and in the future and ML ops is one of our target areas. So, so please, please come back to forebrain again and again, and see what we have to offer.
Ted Hallum: [01:08:18] Awesome. Well, so that's my last question in this section. This is a good segue to it.
So you have, as we said, tremendous perspective, because your background at galvanize you Udacity and now fourth brain I think there's a lot of people out there that are familiar with, or at least have heard of all those programs in terms of outcomes. What, how would the outcomes be different for, or generally be different for people who graduate from fourth brain versus those who get a Nanodegree from you, Udacity or those who go through galvanizes.
Molly Larkin: [01:08:50] Yeah, I can, I can answer that a little bit. I think a lot of it depends on, I think more of it depends on where you want to go and what you want to be. So the question shouldn't necessarily be, should I do a data science bootcamp like galvanize, or should I do a machine learning engineering bootcamp?
Like for sprain, the first question you should ask is what do I want to do in my everyday life, in my everyday professional life? And sort of pick your path based on that. You know, galvanize graduates, our data scientists are, our graduates are machine learning engineers. Of course there's overlap. Of course, you know, there's some companies where that's the same thing and some of our graduates might be going after similar jobs.
So, you know, this is not a clear cut division but the focus areas are different and the skills you learn are a little bit different. So think about what you really want to do. You Udacity you Udacity has great programs and really great content. It's it's an online program that you're doing fully self-directed.
So it's a little bit different than a bootcamp in terms of what you're, what you're getting out of it. And that's not to knock it. I think you Udacity and Coursera of the world have a really important role to play. As you know, for lots of different people, but it's, it's not quite the same intensity and the same purpose as, as a bootcamp style program.
Chris Sanchez: [01:10:18] I actually want to go back one question, Molly and cause, so if I recall, when I remember, I remember when I was first looking at fourth grade and the program that attracted me to it was that it wasn't AI fund company. Right. Because I knew Dr. Andrew Ang is right. I did his machine learning. On Coursera, I did his doing his initial deep-learning course back in 2017, you know, calculating the partial of dilute derivatives of backpropagation by hand doing that.
So I know who he is like he's, he's a Titan in the, in the community, in the data science and engineering community. So I was like, oh, fourth brain. Wow. It's a, it's a, you know, it's this AI fund company. I've got to imagine that they're pretty well networked right within the, you know, in that ship. So I was attracted to that is, you know, getting through it, of being able to get a job at a good company following that.
So I'm going to press you to hear is because not everyone knows who Andrew Green is. But his fourth brain kind of recognized, I guess, in the community is. You know, we're really well renowned company. And, you know, when people are going out and looking for jobs it's is the sports brain something that companies recognize what that is?
Do they recognize this from the AI fund? Or is there an advantage to go into fourth brand if you were looking to get placed at an AI
Ted Hallum: [01:11:53] fund company?
Molly Larkin: [01:11:56] So we, each AI fund company is an independent company. So just to be clear, you know, we're not you know, we're not affiliated officially other than, you know, being part of this AI fund ecosystem does being affiliated with the AI fund ecosystem, a lot of doors open to us.
Yes, absolutely. And you know, we've, we've been lucky enough to get some really, really fantastic partners right off the bat. You know, that some of the companies that came to our very first. Administration day were great companies, you know, Samsung was there, Microsoft was there. Spotify was there representatives from these really great companies.
And, you know, we're, we are a new company. We are a new training program, obviously all that comes with some skepticism as it should. Until, you know, they actually see what our graduates can do, but we got, we got very good back from our, our graduates in our first cohort. And, and many of those companies have committed to come back to, you know, check out our second grad second group of graduates.
So to us, that's, that's a really strong indicator that, that we are, we are in fact meeting the needs of what these hiring managers are looking for.
Chris Sanchez: [01:13:10] Nice. Dang. I missed the boat. I should have done shit.
Ted Hallum: [01:13:16] Well,
Molly Larkin: [01:13:17] we'll get you back in once. Once we have some of our new programs, Chris, I think you'd be a great candidate for them.
Chris Sanchez: [01:13:23] Awesome. So, great segue. It's prediction time when we're talking about machine learning, which is all about predictive analytics. So we're going to do some, some, some fourth brain predictions right now. Right. So we talked about, you know, the evolution of the alternative education field, and I started out in data science and now we're starting to see this focus on deployment, you know, fourth grade being Warren's or being another I'm sure there's a couple others out there.
They're starting to, to focus on that because, you know, as Ted said, employers are starting to really value that skill. Okay. Did it in a notebook. That's great. Now put it up in the cloud somewhere. So an Indian user can actually make use of that model. My prediction three years from now is we're going to start to see the rise of data engineering, boot camps, right.
Where people are going to come in and learn how to build pipelines, maybe learn Java. You know, how did you move, you know, big data from one, you know, one pipeline to another because you know, companies are fighting, Hey, that's great. If we want to do data science, but if our data isn't clean, if we can't get it to move from the point of entry to the actual analysis point, then we're hosed, right.
We might as well, not even have a data science team. So that's, that's my prediction. What do you see? Four fourth brain you'd mentioned as being something coming right now I know you have one program, just, you know, the machine learning engineering program. What are you looking to do to expand over the next three years?
And what kind of challenges. Do you see in, in your particular industry to make that happen?
Molly Larkin: [01:15:05] Yeah, so, so ML ops is definitely a focus area. Something that we we've talked about several times, so I don't need to rehash that, but definitely a, an area that we think students are looking for more specific practice in, and that hiring managers are looking for more specific skills.
And so exactly what that looks like. I'm not quite sure yet, but definitely something that we are working on right now. One of the other things that you know, we see happening, and this is something that, you know, I've, I've heard Andrew talk about is the, in the near future, there will be more off the shelf and sort of accessible and L tools to the, the general population.
And he compared it to Excel in the sense that it's not. It's not that it's simple. It requires some training. It requires some practice, you know, not just anybody can do it, but it is making this, these tools available for a much wider variety of people who don't necessarily know how to code. And so seeing that advance over the next few years, I think is going to lend itself to these, you know, sort of upskilling and training programs.
And Hey, you know, I'm a marketer, I'm a marketing manager, you know, I don't necessarily need an ML engineer to do some of these things that I want to do. I can maybe use some of these off the shelf tools and do it myself. And so I definitely see some of that coming in the next few years as well. What that looks like for a training perspective, I'm not quite sure, but helping, helping people who are non coders and are not trying to be coders they want to stay in their domain area, but they want to leverage these technologies, I think is going to be a new thing in the next few.
Chris Sanchez: [01:16:53] Yeah. And we had talked earlier about, you know, as data science starts to do a better job of defining itself, just like the software engineering community, you know, 20 years ago or so as these roles start to flush out like, oh, a data engineer, does this a data scientist, does this a machine learning engineer?
Does this, do you use brain kind of, I guess, branching out into some of those roles as, as they get better defined? Or do you think that it'll, it'll kind of stick in, in the you know, that, that last half of the stack area, the deployment and the, in the, in the operation,
Molly Larkin: [01:17:28] you mean for fourth brain specifically, or the district in general or both?
That's a great question. I don't know yet. We, we are very, very specifically focused on AI and ML. So obviously all of those roles fall within the larger umbrella, but in terms of making. Sort of enabling more people to benefit from this technology, both from the sense of stable and, you know, exciting employment and also just consumers benefiting from it can look a lot of different ways.
So I'm not quite sure yet
Chris Sanchez: [01:18:07] fair enough. And what to kind of close out what, so what's your biggest challenge right now, right? In terms of growing, you know, growing your your market share or getting, you know, more students involved in the curriculum, is it if I recall correctly, the first cohort you had 20 students, or was it 25
Molly Larkin: [01:18:29] at 25, roughly.
Yeah. And then our, our cohort that, that is in progress right now is 45 and our yup. And our, our, our class that we're launching in may should have between 50 and 50. Depending. So we are we're growing. The, I think that the challenge for any education company is that it's, you know, education in some cases, scales with people.
And so, you know, trying to make sure that we can, can service all the people that are looking for these skills and, and that we are staying current with the, you know, exactly what hiring managers are looking for. That part's easy. You know, staying current is easy. Being able to educate prospective students on what they should have and why they should come to us is, is the challenge for every, every non traditional program out there.
Not just us.
Ted Hallum: [01:19:29] Well, so earlier in the episode Chris, you made mention to that article I wrote, and you cited that statistic about the massive number of people that would need to re-skill by 2025. Just to clarify that wasn't a number I came up with that was from the world economic forum.
And so for anybody who's not familiar with the world, economic forum is a global consortium of the world's thousand largest companies. And so when they put out the 2020 jobs report, which is where that statistic came from, that's basically just a reflection of the holistic perspective of those thousand companies and w what they think they're going to need from a workforce standpoint globally by 2025.
So I've gotten feedback from several people that, all those, those statistics. In like two to two off the charts to be realistic. I'm, I'm sort of taking them at their word given their global perspective that, you know, is, is much higher altitude than than what I personally have. But if they're, if they're right then I think I hope your program is ready to scale massively because it sounds like it's going to be in huge demand and just the next couple of years we're ready
Molly Larkin: [01:20:51] for it.
That's, that's what we, we will be ready for that. Yes.
Ted Hallum: [01:20:55] Awesome. Awesome. Well, you know, as we have veterans in our community, I can't think of a better skill set for them to aspire to. And so I hope for their quality of life and I hope for the United States to be competitive on a global scale with other countries that are going to be fiercely vying with us to.
To fulfill this demand that these companies are going to have for these AI era skills. I hope our veterans listened to you today, thoughtfully. And, and I hope some of them, you know, a decent percentage of them take you up on the offer of the tremendous opportunity, the educational opportunity that's on tap there.
If you could leave our listeners with one parting piece of advice about bootcamps in general, or about your program specifically, what would it be?
Molly Larkin: [01:21:46] I would say they bootcamp, I said this earlier, but I think it's probably the most important piece of advice. If you're trying to decide if it's for you.
You will get out of it, what you put into it. Like if you are really willing to put yourself out there and work hard and think outside the box and kind of just take it all in, you will get a lot of value out of it. You know, regardless of what type of specific role you're looking for, or, you know, what type of skills you're looking for, you can, you can get a lot of value out of a program like this.
If you are willing to kind of put yourself out there and be willing to push yourself outside of your comfort zone. That, and also if you're interested in our program specifically Python to study Python, that's and that's true for almost any AI or ML job or program out there, the more fluent you are with Python, the more comfortable you are with these libraries, the easier it is for you to learn the concepts as opposed to struggling with the syntax.
Ted Hallum: [01:22:55] So my personal preference is for Python. I will say that every professional experience I've had has wanted me to use Python. I have used R in the past. I have nothing against our, I had good experiences with that language. I don't think it is. I don't think it is as marketable of a skill as Python.
And so it doesn't sound like you are either, but I wanted to open up the floor and say, you know, R and Python for people who want, if their goal is gainful employment and marketability. I would say, I think Python wins hands down, hands down. Is that your perspective as well?
Molly Larkin: [01:23:36] I wouldn't say that sign is necessarily more marketable than AR I think in some ways it is, it's more than it's just our focus is the engineering piece.
So knowing the general purpose programming language is important. If you're staying like truly in the analytics side are, are mostly just fine. But that being said, Python is more versatile. So it's never going to be a bad thing for you to learn Python.
Ted Hallum: [01:24:06] I think some people call it a glue language, right.
So if you're having to take analytics and the output of analytics and glue it into a broader ecosystem of things, it makes sense that Python would be your go-to. Yeah. Okay. So a lot of our, I'd say I'd venture to say the vast majority of the people who tune into this podcast are excited about that. It's on some machine learning, they have an insatiable appetite to learn more.
So that means that they look forward to hearing from every guest about other great resources, books, podcasts that are data related. And I definitely want to give you the opportunity to tell us your favorites.
Molly Larkin: [01:24:46] Sure. There there's two newsletters that I read every week and I would certainly recommend the first one is the batch that is from deep learning AI.
Just another one of the, a fund companies and they're many people are familiar with deep learning AI. They offer their courses and specializations for Coursera and they're fantastic programs. The, the batch has an intro written by Andrew in each week. So that's always nice to kind of hear what's on his mind.
Yeah. The second one that I read almost every week is machine learnings with an S the reason I liked that one is because the author leads, but then awesome. Not awesome recap. I, I mentioned earlier, one of the things that we as a team are really interested in is, is ethical AI and, and, you know, increasing representation.
And so keeping, keeping an awareness that, you know, there's some really great advantages of AI and, you know, impacting the world in a positive way. And then there can also be some negative ones. So it's just really interesting to kind of get a quick snapshot of some of those things happening.
Ted Hallum: [01:25:52] Absolutely.
And we'll make sure that links to both of those are in the show notes. So definitely check the show notes. If you want to get keyed in to those two newsletters and get those flowing into your inbox. All right. So Molly, I think the last thing that we have to cover is how people can reach out to you because with everything that you've laid out about fourth brains program I strongly suspect they're going to be listeners who are excited.
They're going to be people who hear this episode and they maybe I've looked at other programs that didn't get into the machine learning engineer piece specifically. And so that is going to excite them about what, what fourth brain has to offer. And they may just want to ask you more questions. May, maybe you've just inspired them and they want to tell you that there's probably a whole laundry list of reasons why people would want to reach out to you.
What are your preferred methods?
Molly Larkin: [01:26:41] You can always reach out to me on LinkedIn. I'm, find-able Molly Larkin working in fourth grade. So always happy to talk to you there. You can also reach out to fourth brain generally on email it's info@fourthbrain.ai. And we'll, we'll come back
Ted Hallum: [01:27:00] outstanding.
I'll make sure that that email address is likewise in the show notes. Okay. With that. Chris Molly, I totally appreciate you guys coming on the show. You've been very generous with your time and to come and share of your knowledge and experiences. I couldn't be more grateful. Thank you so much.
Molly Larkin: [01:27:18] Thank you so much for having me and Chris.
Thanks for joining as well.
Ted Hallum: [01:27:24] Absolutely.
Thank you for joining in to learn more about fourth brain during this conversation with Molly Larkin and Chris Sanchez. Also, I'd like to send a big shout off. Congratulations to Chris who landed a data scientist position at Microsoft. Since this episode was recorded. Great job, Chris, as always until the next episode, I bet you clean data, low P values and Godspeed on your data.