Meta is bolstering its artificial intelligence capabilities with the creation of an elite research team focused on optimizing the algorithms behind Facebook and Instagram. This initiative, detailed in job listings and LinkedIn posts, aims to keep users engaged with content on the company’s core platforms.

Introducing MRS Research

The new team, known as MRS Research, operates within Meta Recommendation Systems (MRS). MRS is responsible for the algorithms that curate the home feeds of Facebook and Instagram users. It also works closely with Meta’s advertising division.

Team Formation and Goals

MRS Research was formally established during an internal reorganization in October. A Meta spokesperson clarified that similar teams have existed in various forms over the past few years. The team’s primary focus will be on long-term AI research and publishing advanced findings to improve Meta’s recommendation engine.

A job description highlights MRS Research as a “newly created org” designed to “leapfrog” existing AI recommendation systems by bringing together leading AI researchers and engineers.

Key Personnel and Recent Hires

Yang Song, Meta’s vice president of recommendation research, leads the effort. He joined Meta in November 2025 after previously leading user growth and recommendations at TikTok, expressing excitement about “revolutionizing” Meta’s recommendation systems.

Shortly after Song’s arrival, Lihong Li, an AI researcher from Amazon, joined MRS Research, as indicated by his LinkedIn profile. Recent additions to the broader MRS division include Xiaolong Wang, formerly of OpenAI, and Fei Sha, from Google.

Meta declined to comment on specific AI talent acquisitions. Attempts to reach Wang and Sha for comment were unsuccessful.

Broader AI Investment

This push to strengthen MRS Research is part of a larger AI talent campaign initiated by Meta in the summer of 2025. This included the launch of Meta Superintelligence Labs (MSL), led by Alexandr Wang, the former CEO of Scale AI. MRS Research operates independently from MSL.

Meta has also recently hired three AI researchers from Thinking Machines Lab, a $12 billion startup. Internally, the company has rebranded employees as “AI builders” and encouraged them to experiment with AI-driven product development.

Meta has been actively exploring the use of AI to enhance its algorithms, launching an AI model in late 2025 that reportedly improved ad performance by delivering more relevant ads to users.