Ricardo Olmedo

Also published as Ricardo Dominguez‑Olmedo

I am a final-year PhD student at the Max Planck Institute for Intelligent Systems, Tübingen, working with Moritz Hardt and Bernhard Schölkopf.

My research focuses on LLM benchmarking & specialization.

During spring/summer 2026, I visited Sanmi Koyejo at Stanford University. My PhD is supported by a Google PhD Fellowship (2025–2027).

Selected works

All publications →

nanoswe: Speedrunning SWE-bench In submission
Training on the Test Task Confounds Evaluation and Emergence ICLR 2025Oral
Lawma: The Power of Specialization for Legal Annotation ICLR 2025 + CS&Law'25

Blog posts

Sep 2025 mini-coder: small models for research on SWE agents
May 2025 How to Fix the LMArena? Release All Data

Recent talks

Oct 2026
Speedrunning SWE-bench, or the $1,000 coding agent Confluence Workshop @ EPFL
Sep 2026
Benchmarks and How They Drive Progress in AI Machine Learning Summer School, Tübingen
Jul 2026
Scaling, specialization, and speedrunning SWE-bench SALT & Talk, Stanford University