How to build a side-project to get a job in Machine Learning [Storytime Saturdays]
The most important steps to highlight your skills and get your first AI job.
Hey, it’s your favorite cult leader here 🐱👤
On Saturdays, I will cover stories/writeups covering various people’s experiences 📚📚. These stories will help you learn from the mistakes and successes of others. These stories will cover various topics like Leadership, Productivity, and Personal/Professional Development. Use these to zoom ahead on your goals 🚀🚀
To get access to all my articles and support my crippling chocolate milk addiction, consider subscribing if you haven’t already!
p.s. you can learn more about the paid plan here.
So how can you develop your career in Machine Learning?
A lot of people reach out to me with this question. Answering this question is complex and relies heavily on the person’s individual goals, interests, and skills. For example, while I am good at interacting with AI Research- I didn’t like being in college (or any schooling whatsoever). Which is why I never tried to pursue an upper-level degree. This decision means that I am not eligible to apply for research roles at Big Tech Companies, but this is a tradeoff I’m willing to take. Other people really want to work in Amazon’s AI Lab- in this case getting an advanced degree is non-negotiable.
Other factors such as your location, prior experience and your networks further complicate this question. Therefore, it would be impossible to give you a blanket piece of advice that would go beyond generic platitudes. However, we can do better if we break it down by various demographics. In this article, I will focus my advice on the group that I am most qualified to speak to- undergrad students looking for their first role in Machine Learning. Not only was I once in this position, but I’ve also worked with people in similar positions.
The most crucial step to getting your first ML job is to have an amazing side project. So that is what we will be covering in this article. Scrap the boring Iris datasets, no more Titanic.csv- this piece will cover how you can build truly exceptional side projects that will help you stand out. Sound exciting? Let’s get right into it-
PS- If you’re a software developer looking for more career advice, I would highly recommend subscribing to the free newsletter . is a staff software engineer at Instagram, helping close the gap for engineers without mentors. He writes snappy articles filled with career advice.
The Ingredients to an Epic ML Project
The biggest mistake people make- A mistake I see people make repeatedly is to spend too much time/energy emphasizing the model they built. They’ll spend hours tweaking model architectures/running hyper-parameter sweeps. Unless you can jump to the top of Kaggle competitions or SOTA benchmarks- this isn’t useful to anyone. Here’s the truth, for most projects models don’t really matter. You’re probably not doing something amazing or groundbreaking in AI Research All that energy wasted in trying to hit peak performance is better spent building other parts of the project. Here’s what you should do instead.
Index heavily on data engineering- Data Engineering has a few advantages: no master’s degree requirements, high demand, and ability to transition internally to more Data Science/AI focused roles. Therefore, it makes sense to develop your projects to heavily showcase your Data Engineering skills. This will leave you eligible to both DE and MLE roles (which is not something you would have if you spent most of your time tinkering with the model). Building your base in Data Engineering will help you a lot when you start scaling up your systems and will give you a huge advantage over ML Engineers who come from a research background (since most research doesn’t really focus on deploying products and/or scaling up pipelines).
Take Data from a local business/non-traditional source- This is how I developed my skills in AI. I would go up to local businesses, tell them I would help them with AI, and offer to build MVP solutions for free (as a proof of skill). If they liked my work, we could talk about paying me. 90% of the time you will end up doing free work. This is fine- because this becomes your side project. In this case, you will learn all kinds of valuable skills- setting up applications, handling data from various sources etc. These will be unbelievably valuable in your interviews and applications. PS- I’m not advocating for unpaid internships. In this case, you will be the one calling the shots, setting your schedule, and deciding what to do. This is just a side project that a local business evaluates. Working with local businesses also has other benefits: your dataset and challenges will be unique; you will have more networking opportunities; and you get to see what works IRL and what doesn’t. And in case the business likes you enough to hire you- congrats you got your first job in AI. This might not be glamorous, but real work experience with demonstrated results goes a long way to getting big boy jobs.
Focus heavily on visualizations and demonstrations- This is straightforward: show don’t tell. Simple visualizations and easy to use apps make things easier for the end user- which is always a huge plus. People will give you higher marks for graphs and interactive charts, even if these are relatively basic.
Build complete projects- Similar to the last point- don’t just leave your project as a bunch of code to be compiled. Take the time to turn it into an executable or application. Not only is this is a great skill (cloud/remote deployment is necessary for modern ML Practitioners), but it will also make things easier for someone who wants to check out your demos. Doing that also adds a bit of polish to your overall project and makes you look more professional.
Package up the project with YouTube videos + Design Documents- This is a bonus tip. When I would make projects, I would record YouTube video demos going into depth about what I did. This has a huge advantage: you get a chance to demonstrate your knowledge. The person looking at your final project doesn’t know all the experiments you ran, policies you tested etc. By having YouTube videos about it, you can really show off your knowledge/detail your experiences. People are sometimes hesitant to go to your website/download apps- but they will be less worried about looking through a YouTube channel.
It’s important to remember that most hiring managers will not look at your channel. But the ones that do will respond very positively (multiple hiring managers have given me job offers/referrals based on my YouTube channel, even though it’s nowhere as polished or prominent as my writing). Well-structured design documents will also be a powerful addition since that is a huge skill for engineers. Finally, Design docs and YouTube will teach you how to communicate better, just blasting you with benefits all over.
As a general principle- projects should mirror real-world products and address similar challenges. Effective ML Engineering is a combination of Theoretical AI, Software Engineering, and System Design. If you are an entry level developer, you can add a lot of value by focusing on the latter 2, since most AI Projects are not cutting-edge models but rather simple models applied in various business contexts with specific challenges. For example, this is the model that Amazon uses for dealing with their multi-billion-dollar bot-detection problem. While there is some work in the data prep/augmentation, the model is fairly straightforward. Most hiring managers/teams would benefit more from an intern/junior who can write scalable ML software compared to one who knows the most cutting-edge models/protocols. Develop your projects accordingly.
Loved the post? Hate it? Want to talk to me about your crypto portfolio? Get my predictions on the UCL or MMA PPVs? Want an in-depth analysis of the best chocolate milk brands? Reach out to me by replying to this email, in the comments, or using the links below.
Reach out to me
Use the links below to check out my other content, learn more about tutoring, reach out to me about projects, or just to say hi.
Small Snippets about Tech, AI and Machine Learning over here
AI Newsletter- https://artificialintelligencemadesimple.substack.com/
My grandma's favorite Tech Newsletter- https://codinginterviewsmadesimple.substack.com/
Read my articles on Medium: https://rb.gy/zn1aiu
My YouTube: https://rb.gy/88iwdd
Reach out to me on LinkedIn. Let’s connect: https://rb.gy/m5ok2y
My Instagram: https://rb.gy/gmvuy9
My Twitter: https://twitter.com/Machine01776819