An AI researcher at Meta making over $250k said publishing research helped her land the role

This as-told-to essay is based on a conversation with Ruiyu Li, a 26-year-old AI researcher at Meta, based in Menlo Park. Her identity, employment, and salary (which includes base, bonus, and equity) have been confirmed by Business Insider. This story has been edited for length and clarity. I'm a research scientist at Meta specializing in AI and machine learning.

I work on training AI models to advance large-scale recommendation systems, building AI-native agents to improve efficiency, and optimizing LLM post-training and inference. I graduated from college in 2022 and finished my Master's in machine learning at Carnegie Mellon University in 2023. It was a difficult time in the tech job market.

I found career fairs and conferences useless — but I eventually landed three Big Tech offers. I went to Microsoft first, but later ended up at Meta. That experience gave me practical lessons on what actually helped to get tech offers during a difficult market. If you want to work in AI, you need to be familiar with algorithms and coding.

AI is helping people do a lot of the job, but the foundational knowledge is still necessary. You should also have deliverables to show the employers. I encourage people who want to enter the field to gain as much hands-on AI experience as possible so they can adapt as the field evolves rapidly. I recommend either interning at a startup or publishing research.

I interned at a startup in college and at Microsoft during grad school, and both required pushing AI innovations into production. When I interned at the startup, there were fewer than 10 people, and I worked directly with the CEO and CTO. When I look at candidates, I check if they have research and projects.

During competitive job markets, it's not enough to simply list experience or projects on your résumé. If you have published papers or an open-sourced codebase that employers can look at, that's more helpful. Before joining Meta, I conducted academic research at eight AI labs across several areas of machine learning.

At Meta, I've applied that foundation to production AI systems, where model quality must be balanced with scalability, reliability, efficiency, and measurable real-world impact. The more you do, the more practical experience you gain, and the quicker you can learn and adapt to AI trends. If you want to work specifically in AI, I recommend going to grad school.

I decided after college that I wanted to work more on applied AI research. My bachelor's degree was in computer science, but I wanted to go more in-depth. My master's program had a lot of advanced projects that were practical, and it allowed me to be a part of frontier research. My internship experiences were also very important in landing a full-time job.

Many employers want to see tech experience and solid industry-level projects. When I interned at a startup, I had the chance to work directly on video-editing automation using AI innovations, which were state-of-the-art at the time. During my internship at Microsoft, I worked on generative search and Bing Copilot, and delivered an end-to-end video generation pipeline using a multimodal LLM that was shipped to production at Bing Search.

In both internships, I developed a niche by focusing on large-scale AI systems in production environments and AI video generation. I developed many transferable skills and experience from my internship that carried over to my full-time role. Do you work in Big Tech? We want to hear about how you got your job.

Reach out to the reporter at [email protected].