Researcher, Alignment CoT Monitorability
- Location
- San Francisco (United States)
- Workplace
- Not stated
- Type
- Full-time
- Team
- Research
- Salary
- USD 295,000 – 500,000
- Posted
- 2026-08-04 (3mo ago)
- Source
- OpenAI's career page (Ashby)
- Last checked
- 2026-11-10 (1h ago)
About the role
About the Team The CoT Monitorability team at OpenAI studies whether and when the chain-of-thought of frontier reasoning models is monitorable enough to support scalable oversight. We study how to measure monitorability , which training mechanisms affect monitorability, and speculative methods to improve monitorability. While we mostly focus on CoT monitorability at the moment, we care more generally about any form of monitorability, auditing methods, and improving alignment.…
Read the full posting and apply on OpenAI's own site
A free account opens 5 complete job posts, apply links included, and searches every role we index by title, company, country, workplace, level and pay. No card needed.
Create a free account Log inMore open roles at OpenAI
- Account Associate - Singapore (Bahasa-speaking) · Singapore · Hybrid
- RTL Design Engineer, Networking / NoC · San Francisco, United States · Hybrid
- Principal Software Engineer, Full Stack — Codex · San Francisco, United States
- Program Manager, M&A Recruiting | People · San Francisco, United States · Hybrid
- Recruiter, Data Center Infrastructure · San Francisco, United States · Hybrid
- Android Engineer, Growth · San Francisco, United States · Hybrid
Details come from the employer's career page and may change. Confirm them on the employer's site before applying. Report or request removal · How JobsScanner works