Mohamed A M Elansary, PhD
Research Scientist, Frontier Capabilities — Lila Sciences
Expertise Area 1 — Agentic system building
Why this seat
- Agentic Area 1 — production agentic LLM systems (investigate → synthesize) for long-horizon scientific workflows.
- Scientific regime measurement — PhD hard-eval / UQ under delayed feedback and contested multi-source ground truth.
- The PhD who ships — baselines, ablations mindset + shipping, not claim theater.
Sourced insight 1 — JD
“Feedback is sparse and delayed — experiments take days or weeks, not milliseconds. Ground truth is expensive or contested. … Existing benchmark do not capture these nuances.”Source: Greenhouse · RS Frontier Capabilities · 2026-09-03
Sourced insight 2 — Company
“Verifiable rewards have driven AI breakthroughs in math and coding, but most of science is hard to verify. Lila's AI Science Facilities Factories change that… the world's largest verifier for science.”Source: lila.ai/tech · 2026-09-03
Honest gaps
- No claimed Lila AISF / therapeutics / materials ownership.
- Verified pubs only — nothing invented.
- Differentiated from sibling Applied AI MLE pack.