Fred Sala

Chief Scientist · Snorkel AI

Chief Scientist at Snorkel AI and assistant professor of CS at UW-Madison; drives Snorkel's benchmark and evaluation research (OSWorld 2.0, Senior SWE-bench, SlopCode Bench, RIFT) and sits on the Open Benchmarks Grants steering committee.

medium confidenceMadison, WIUpdated 2026-09-19

Background

Researcher on the fundamentals of data-driven systems, data-centric AI, foundation models and automated ML; 2024 DARPA Young Faculty Award, NSF Graduate Research Fellow. His UW-Madison group co-develops several Snorkel-backed benchmarks.

What they run now

Benchmarks and evaluation methodology for agents: OSWorld 2.0 (Jun 2026), 'Can Generalist Agents Automate Data Curation?' (Jun 2026), 'Models Can Model, But Can't Bind' (COLM 2026), RIFT rubric taxonomy (Apr 2026); steering committee of the $3M Open Benchmarks Grants.

Career

presentChief Scientist, Snorkel AI
presentAssistant Professor, Computer Sciences, University of Wisconsin-Madison

On the record

Benchmarks · 2026-07

Says the grant projects target the field's hardest evaluation challenges, as AI capability outpaces measurement. source

Interviews & talks

Chat with the Terminal-Bench team — Snorkel blog/webinar
Applying information theory to ML with Fred Sala — Snorkel talk

Connections

Paroma Varma — co-author
Vincent Sunn Chen — colleague on Open Benchmarks Grants

Why it matters here

If a visual/design-judgement benchmark were ever funded by Snorkel, he would be one of the reviewers. Moderate indirect relevance.

How to reach

Academic route: UW-Madison; responds to research collaboration rather than vendor pitches (speculative).

What we could not establish

Start date at Snorkel
X handle @fredsala appears to be his but bio/company link not verified
Education details

Sources

Also at Snorkel AI

high confidence

Fred Sala

Background

What they run now

Career

On the record

Interviews & talks

Connections

Sources

Alex Ratner

Chris Ré

Henry Ehrenberg

Paroma Varma

Vincent Sunn Chen