Miju Labs

← All buyers · Meta (Meta Superintelligence Labs)

N

Nat Friedman

Head of Products and Applied Research, Meta Superintelligence Labs · Meta (Meta Superintelligence Labs)

medium confidenceTier B · Map and watchFrontier labsimage · video · text

Former GitHub CEO, now head of Products and Applied Research at Meta Superintelligence Labs since June 2025, reporting to Chief AI Officer Alexandr Wang. He owns the consumer-integration side — the surfaces where Meta's Muse image and video models actually reach people — which makes him the nearest thing Meta has to an executive sponsor for output quality that users judge by eye.

Background

Friedman ran GitHub as CEO after Microsoft's acquisition, then with Daniel Gross built AI Grant and an investment practice that made them two of the better-connected people in AI. Both joined Meta in mid-2025 — Friedman on 30 June 2025 as head of Products and Applied Research, Gross in July — as part of the Superintelligence Labs formation, reporting to Alexandr Wang.

MSL is organised into four groups: TBD Lab (frontier models), FAIR, Products and Applied Research (Friedman), and MSL Infra (Aparna Ramani). Friedman's group is where model capability becomes a Meta product, which for images and video means Meta AI, and the Muse Spark, Muse Image and Muse Video line. He is a product-taste person by reputation, and his investing history skews toward developer and creative tools.

What they run now

Turning MSL's models into Meta products: Meta AI and the consumer surfaces that carry image and video generation. In practice he owns the perceived quality of what users see, which is the applied end of the taste problem.

What they buy

Inference. A products-and-applied-research head buys evaluation and user-facing quality measurement more than training data: does the generated image look good enough to send to a friend, does the video hold up in a feed. Possible objects: consumer-facing aesthetic preference studies across demographics and cultures, brand-safety and quality gates, and creative-professional panels for the pro-adjacent use cases. Meta's Scale AI relationship covers much of the generic labelling need, which narrows any external opportunity to specialist judgement.

Career

  1. 2018 – 2021CEO, GitHub
  2. 2021 – 2025Investor, AI Grant / Nat Friedman & Daniel Gross
  3. 2025-06-30 – presentHead of Products and Applied Research, Meta Superintelligence Labs — Reports to Alexandr Wang
  4. —Massachusetts Institute of Technology, BS, Computer Science and Mathematics
Why it matters here

Medium as a sponsor, low as a realistic contact. He is the right altitude and the right sensibility — a product-taste person over a consumer creative surface — but Meta is the least accessible of the four labs for an outside data vendor and he is deep inside a large organisation. Keep him in the index as the escalation target if a Meta conversation ever starts elsewhere.

How to reach

Not a first target. Speculation: he is unusually reachable for someone at his level given his investor background and public presence, and would respond to a genuinely novel finding about creative-output quality rather than to a services pitch. Any approach should assume he knows the data-vendor landscape better than most buyers do, because he invested in parts of it.

What we could not establish
  • No verified X handle or LinkedIn captured in this pass.
  • Whether Muse Image and Muse Video product ownership sits in his group or in TBD Lab.
  • Meta's external human-data procurement beyond Scale AI is not public.
  • Education listed from general public record, not confirmed here.

Why they are on the map

Leads the product arm that ships Meta AI and Vibes features. Role confirmed Aug 2025; not confirmed in 2026. source — evidence dated 2025-08

Decides whether the problem is funded.

How Meta (Meta Superintelligence Labs) buys data

Aug 2025: TBD Lab researchers reportedly preferred Surge and Mercor data over Scale's, despite Meta's $14.3B Scale stake. Meta has also partnered with Mercor on RL environments (TechCrunch, Sep 2025). In Aug 2025 Meta licensed Midjourney's 'aesthetic technology' to make its models 'more visually appealing', which is a direct signal of paying for visual taste. Muse Image/Video (Jul 2026) are marketed on human-preference Elo from internal ablations and Arena. SemiAnalysis (Jul 2026) reports about 3,000 engineers working on RL task/environment creation.

Alexandr Wang (Chief AI Officer, org context only) heads MSL. Since Aug 2025 it has four parts: TBD Lab (frontier models), FAIR (Rob Fergus), Products and Applied Research (Nat Friedman) and MSL Infra (Aparna Ramani, Amir Frenkel). Shengjia Zhao is Chief Scientist and owns the Muse roadmap; Jiahui Yu leads multimodal, including Muse Image/Video. No human-data owner is public; data-vendor decisions appear to sit with TBD Lab researchers and ex-Scale staff.

Known vendors: scale-ai, surge-ai, mercor, lmarena, Midjourney (aesthetics tech licence)

Sources

  1. Meta Superintelligence Labs
  2. Nat Friedman
  3. Meta Superintelligence Labs: What We Know So Far
  4. Meta's AI Reorg