- Saurabh Khandelwal started his machine learning career at a startup before moving to Big Tech.
- He highlighted specific experiences in both startup and Big Tech to enhance his résumé.
- His résumé's overall narrative helped show that he understands machine learning systems end-to-end.
This as-told-to essay is based on a conversation with Saurabh Khandelwal, a 28-year-old machine learning engineer at Meta, based in Bellevue, Washington. The following has been edited for length and clarity.
There are so many different paths within the world of AI and machine learning.
For people getting into the field, it's important to be deliberate with how you spend your time learning. I've worked in both startups and Big Tech companies, and I've had to be intentional about how I present my career. I'm now a machine learning engineer at Meta.
The technology will keep changing, and new models will emerge, but it's important to focus on the why of what you're doing for both the work itself and how you present yourself as a difference maker in that work.
When I applied for my role at Meta, I focused on building out a narrative throughline in my résumé, which I think helped my application and brought me closer to my long-term career goals.
I started within the startup world before joining Big Tech
My first job was as a founding machine learning engineer at a startup. Then, in the summer of 2022, I was interning at a hedge fund in New York as a software engineer, and I applied for a role at Amazon.
I joined Amazon in February 2023. I really enjoyed my time there. It taught me a lot about how Big Tech approaches problem-solving and scaling resources.
I was finishing my existing projects at the end of 2024 and wanted to get into a machine learning role that focused on both research and implementation, since my role at Amazon was more limited to just implementation. I started looking into opportunities and, by the end of February 2025, left Amazon to join Meta as a machine learning engineer.
I have more independence at Meta
My role at Meta is more closely tied to product outcomes than what I did at Amazon, which is part of why I was interested.
At Meta, I focus on both research and implementation. There's less focus on getting alignment from everyone on my team than in my previous role at Amazon. If I feel an idea is good, I'm encouraged to try it, test it, and ship it to production with the necessary guardrails in place.
I built a narrative in my résumé to help me get the job at Meta
When I was applying for this role, I didn't just list my projects on my résumé; I focused on highlighting details that show the strongest throughline. I wanted it to tell a story.
My narrative was two parts. The first was that I understood machine learning systems as a whole, because I worked at a startup before joining Amazon, and the scope of tasks at a startup is way broader than at a Big Tech company. Everything is built from scratch, and I had to understand each and every process step by step and build components accordingly.
The other part of my narrative was that I can solve machine learning systems at scale because of my experience at Amazon. On my résumé, I was specific about the scale I worked at, how many tokens I processed, and how many requests I serviced, because Big Tech companies care about systems that operate at that level. Even startups care about those scaling data points because, at some point, they want to be operating and deploying at that level.
The overall narrative was that I understand the system end-to-end and can be a key difference-maker in shipping one idea to the final destination because I understand the full life cycle of the machine learning system.
My best advice for machine learning engineers
Machine learning is moving so fast right now that it's hard to keep up.
If you are a bit more experienced in the industry, know which system or part of the problem you want to focus on, and really follow that research. Follow bloggers, go to conferences in the specialties you're interested in, and go deep on those problems. I have a dedicated time for learning every week, during which I block my calendar for one hour.
If you are just starting your career, focus on having a strong foundation in the base on which machine learning systems are built. If you understand the basics well, you can adapt and deploy these systems even as architectures and tools change a hundred more times.
Do you have a story to share about your career journey in tech? Contact this reporter, Agnes Applegate, at [email protected].
