Home People Organizations Reports Fields Schools

Public Atlas People first, reports as evidence, organizations as context.

Atlas / People / Detail

Mrinank Sharma

AI safety researcher who led Anthropic's Safeguards Research Team and worked on jailbreak robustness, automated red teaming, and monitoring for misuse and misalignment.

Researcher1 organizations1 reports

Profile status: updated

Suggest a correction

Suggest a source

Trust signals

Profile completeness43%

Public sources2

Official sources0

Last reviewedMar 13, 2026

updated 2 public sources

Public links

news Anthropic team announcement x Public X profile

Organizations

core Anthropic

Reports

Alignment and Safety Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Training

Supporting sources

https://alignment.anthropic.com/2025/introducing-safeguards-research-team/ Supporting source · news https://x.com/MrinankSharma Supporting source · social

LLMpeople is a public atlas for discovering frontier AI researchers with context, provenance, and respect.

Privacy · Terms