How I Became a Machine Learning Practitioner

https://blog.gregbrockman.com/how-i-became-a-machine-learning-practitioner

by sama • 7 years ago

290 47 7 years ago

blog.gregbrockman.com

minimaxir• 7 years ago

Given that Greg Brockman was the CTO of Stripe before OpenAI, that's a order of magnitude more technically/CS capable than the typical reader who might be looking into ML.

copperx• 7 years ago

Studying Math at Harvard/MIT certainly puts you in a different category than the average software engineer. And if ML was still challenging to Greg, it is honestly a bit discouraging.

gdb• 7 years ago

(I wrote the post.)

If it's helpful, I dropped out of both schools — the vast majority of my knowledge is self taught!

streetcat14• 7 years ago

Maybe if you want to write tensorflow, but not if you want to use it.

omarhaneef• 7 years ago

I think "doing" ML math is a little different than "doing" math math.

With math, on paper say, it is hard to tell if you are doing it right or wrong. You can trick yourself quite easily. A compelling proof can have a huge hole.

You can still trick yourself programming -- in a sense, that is what a bug is -- but it is much harder.

The upshot is, I think it is easier to teach yourself math that is applied to a computer program than math on a piece of paper.

dhairya5• 7 years ago

So there is a difference between ML research and application. Being a practitioner doesn't require deep math knowledge that perhaps research would. Jeremy Howard's fastai course is a great example of how someone with a solid programming background can effectively transition into being a deep learning practitioner. Given that production ml and deep learning is still the wild west, as a practioner, you can contribute also to the research around effective training, scaling, and application of these models. The math and intuition required are definitely acquirable.

I think when you shift into pure research, yes a deep probability, information theory, linear algebra, and calculus background are needed. But at the level, you're rarely writing code and more likely working at theoretical level.

dhairya5• 7 years ago

Yes but the transition definitely doable and his advice is great. The key part of his advice is spending time experimenting, rapidly failing, and continuing to work on it with real world use case. Often the challenge is making the jump from the simple toy examples used in educational materials to the messiness of real-world data.

I'm a senior data scientist at vc-back startup. I'm in a hybrid data scientist/ machine learning engineering role, where I build and train ml and deep learning models and also build the scaffolding them to support their production usage. But my previous roles included being a business analyst, project manager, and research analyst. My undergrad education was in Creative Writing and the social sciences.

While I kind of accidentally transitioned into this career, how I got here is similar to most folks coming from a different background. Lot's of self-study and experimentation. I think one of the challenges to transitioning into ML and deep learning is that there are so many applications, domains, and input formats. It can be overwhelming to learn about vision, nlp, tabular, time-series and all other formats, applications and domains.

Things solidified for me when I found a space I found compelling and I was able to dive deep into it. You kind of learn the fundamentals along the way through experimentations and reflection. My pattern was pick up a model or architecture. Learn to apply it first to get familiar with it, experiment with different data, and then go back to build it from scratch to learn the fundamentals. That and I read a lot of papers related to problems I was interested in. After a while, I started developing intuitions around classes of problems and how to engage them (in DS you rarely ever solve the problem, there's always room to improve the model ...)

streetcat14• 7 years ago

Thanks for the info.

I have a serious question (not for bashing)

Can you please describe what part of your job CANNOT be automated?

mastratton3• 7 years ago

This is really good advice. I'd say it generalizes to learning most engineering challenges: Pick a problem you're interested in and solve it from top to bottom, tweaking all components as you move forward.

sgt101• 7 years ago

We know that someone with a good CS degree can do this because.. Ph.D students....

jorblumesea• 7 years ago

Does anyone not want to become a ML engineer? Is this the future, and will we even have a choice or else be out of a job?

frenchman99• 7 years ago

Well, you could save money and use it to feed yourself while learn new skills when you're out of a job. That works too.

klipt2• 7 years ago

There's plenty of non ML software engineering to be done. Anecdotally a large proportion of interns want an "ML project", but only a small percentage of teams looking for interns are offering one.

Too many people going into ML could skew the supply/demand into making it a worse job option (more work, less pay), like game programming or academia.

seanmcdirmid• 7 years ago

There aren’t that many engineers with 10+ years of ML experience ATM, but there will be tons in 5 or so years. Chasing tomorrow’s trend is always more productive than chasing today’s, but of course the former requires predicting the future.

rhacker• 7 years ago

The question is this: 10 years from now when the top job requirements list ML - are you going to be ready or out of the game?

threeseed• 7 years ago

I started in ML about 7 years ago so well before the hype and back then very few people wanted to be ML engineers.

What's happening at least in Australia now is that contract rates (a good indicator of the supply/demand ratio) has halved for ML engineers. Which means (a) a lot of people want to be ML engineers and (b) there aren't that many jobs for them.

thundergolfer• 7 years ago

Just to be sure I understand correctly, you’re saying that money has halved for ML engineers because plenty of people are getting into ML and are willing to take much less money? You said they “want to be” ML engineers which implies they aren’t really and thus aren’t worth ‘full-rate’

bitL2• 7 years ago

ML engineer is a super boring job content-wise and has insane outside pressure. It's about building data pipelines, the ugly grunt work. ML/Data Scientist is the interesting job. Usually Data Scientists view ML Engineers as replaceable drones that don't understand anything interesting and do the boring part of the job for 2-3x less than they do. The only advantage of ML Engineers is that AutoML is unlikely going to replace some dirty work but might endanger outdated Data Scientists.

nilkn• 7 years ago

I might disagree on this. The software engineering behind production machine learning systems can be quite interesting and nontrivial. It really depends on the scope of the challenges being faced. If you have thousands of models that need to be served in production and continually retrained and monitored, that becomes a pretty sophisticated problem space to work in.

whymauri• 7 years ago

Except for the part where data science is an incredibly broad term and the majority of the positions are seemingly what used to be called 'data analyst'.

It makes finding good positions really hard.

anon44• 7 years ago

the vast majority of ML/data scientist positions are actually similar to the hypothetical ML engineer position you described.

badpun• 7 years ago

I got an offer to do ML on video data recently, but turned it down to do some non-ML software eng. instead, for double the money. The ML work would be way cooler, but it would at a startup (=stressful) and double the money means half the time to reach FIRE.

The caveat is that I've worked on ML in the past and I think that the work is maybe less intellectual than software engineering - with complex enough models they become impossible to understand and you start to just try out ideas based on random intuitions. The thing I mostly like about it is the ability to use math and more independent style of work - no scrum, less need for cooperation with other team members etc..

mychael• 7 years ago

Congrats on your cool life, your ivy league education, your CTO role at OpenAI and all the access that provides. You've done it!

Also thanks for telling us how you became a practitioner. It's definitely relatable and not a humble brag at all.

ilyasut• 7 years ago

Comments like this are what make HN great and are not at all toxic!

strikelaserclaw• 7 years ago

Jealousy is unbecoming. I had no idea who this guy was and i found the article nice to read. It is a story about a guy who feels out of his depth in a highly technical domain but instead of giving up, he devotes time to it and comes out the other end much more competent. The internet provides everyone with world class experts to help you if you are stuck on something, most people don't have the will to help themselves.

TrackerFF2• 7 years ago

I think it's doable if you're at the right place and have enough opportunities around you, and something to show for. I feel that companies and startups around tech hubs are more willing to give someone a chance, and look through the formality, if you manage to convince / impress them.

Where I live, far away from tech, it's almost impossible to land a job in ML / AI / DS unless you have a (minimum) Masters degree in something relevant. Preferably a Ph.D and solid experience to show for - I know because I work in the field, and lots of F500 dinosaurs are just now waking up. But are also unfortunately clinging to their old ways of hiring people.

Schools all over are also picking up slack, starting to offer specialized graduate degrees in those domains. When I got my degree in ML, it was a sub-field at my schools engineering department, mixed up with signal processing and control theory groups.

When first trying to get a job, the main problem was to explain what I actually could bring and do, and a lot of the recruiters or managers had no idea what Machine Learning was. Then you said "It's basically Artificial Intelligence" and, and they were instantly wooed.

rolltiide• 7 years ago

Your observation is correct. No degrees are needed there is a lot of stuff to build.

croh• 7 years ago

Indeed inspirational article.

How I Became a Machine Learning Practitioner

Related Stories