Member of Technical Staff - Research, AI Safety
Vals AI · San Francisco
Applied AIonsite$130k–$185k
At a glance
Highlights
- Design next-gen benchmarks to catch reward hacking, scheming and deception.
- Build evaluations that shape how frontier labs develop foundation models.
- Publish safety research and define the standard for real-world LLM evaluation.
What they offer
- Meaningful ownership
- Relocation and transportation support
- Health and dental insurance coverage
- Lunch and dinner provided plus free snacks and drinks
- Unlimited PTO
- $1,500 housing stipend within one-mile radius
- Collaborate with leading AI labs and publish research
Ideal candidate
Deep AI safety researcher who loves building rigorous evaluations and seeing work shape frontier models. Fast-learning, high-ownership builder eager to work in-person on high-impact misalignment problems.
Requirements
- Substantial hands-on work in safety evaluations, monitorability, scheming, deception or multi-agent safety
- Deep ML/AI background with understanding of neural networks and model training fundamentals
- Excellent coding skills to design and build complex evaluation environments in Python
- Ability to clearly communicate complex research ideas to technical and non-technical audiences
- remote policy
- In-person only
Lab profile
About Vals AI
Builds expert-driven benchmarks and platform to evaluate LLMs for safety and enterprise use.
What they build
- LLM evaluation platform
- AI safety benchmarks
- Cybersecurity benchmarks
Known for
- LLM evaluation at scale
- AI safety benchmarking
- Cybersecurity benchmarking
- Expert-led scenario design
Posted · Last seen