Joel Becker

I'm currently trying to figure out what on earth is going on with AI recursive self-improvement.

Previously, I was Founder/CEO of Qally's and did economics/genomics research at NYU, Oxford, and Harvard.

Outside work: I was once the most profitable forecaster on Manifold Markets, co-ran a statistics consultancy advising professional 'soccer' teams, and built a social media presence peaking at 300,000 followers. I enjoy a good sing-along, keeping physically (and financially!) healthy, and cultivating my digital garden.

Email: joelhbkr@gmail.com

Anonymous Feedback: form

Research

Frontier Risk Report (February to March 2026)

A Cotra et al. (2026), METR Blog Post

Many SWE-bench-Passing PRs Would Not Be Merged into Main

P Whitfill, C Wu, J Becker, and N Rush (2026), METR Blog Post

We are Changing our Developer Productivity Experiment Design

J Becker, N Rush, T Cunningham, D Rein, and K Mahamud (2026), METR Blog Post

Forecasting AI Time Horizon Under Compute Slowdowns

P Whitfill, B Snodin, and J Becker (2025), Working Paper

Measuring the Impact of Early-2025 AI on Experienced Open-Source Developer Productivity

J Becker et al. (2025), Working Paper

Measuring AI Ability to Complete Long Tasks

T Kwa et al. (2025), NeurIPS 2025

HCAST: Human-Calibrated Autonomy Software Tasks

D Rein et al. (2025), Working Paper

RE-Bench: Evaluating frontier AI R&D capabilities of language model agents against human experts

H Wijk et al. (2024), ICML 2025 (Spotlight)







Contact

Email: joelhbkr@gmail.com

Anonymous Feedback: form