FIRST SEENJUN 21, 2026RUNREQ OBSERVATIONDAYS OPEN5744D COMPANY MEDIANTEAM OPEN66AI RESEARCH & ENGINEERING30D TEAM+6NEWLY OBSERVEDTRACKED COMP$350K–$850K /YRDISCLOSED RANGE / ALL TIERSSOURCECLOSEDCHECKED 13D AGO
JOB DESCRIPTIONSHORTENED FROM SOURCE
Key responsibilities
4 ITEMS
Lead research projects that detect Claude misuse, malicious accounts, model safety risks, and abusive behavior.
Analyze model usage data, build detection systems, and evaluate safeguards in chat based and agentic workflows.
Develop prototypes for real time safeguards and transfer successful methods to production engineering teams.
Build evaluations, measure safeguard effectiveness, and communicate findings to Trust and Safety, research, and product teams.
Qualifications needed
4 ITEMS
Have a Bachelor's degree or equivalent education, training, or experience in a relevant field.
Independently drive research from ambiguous problems to results in artificial intelligence, machine learning, security, integrity, or related fields.
Scope own work across research, engineering, analysis, and understand large language model sampling, prompting, and training.
Use Python and work with large datasets.
EVENT>Anthropic :: Research Engineer (Senior Staff+), Safeguards Labs first observed 57 days ago