The Interpretability team at Anthropic focuses on reverse‑engineering how large language models operate to enable safety and reliability. This role involves building and scaling the inference and training infrastructure that supports research tools, profiling and optimizing large‑scale distributed systems, and translating research needs into production‑ready engineering solutions. Candidates should have extensive software engineering experience, be proficient in Python and at least one other language, and be passionate about AI safety and interpretability. The position is based in the San Francisco office with the possibility of remote work on a case‑by‑case basis.
Research Engineer, Interpretability at Anthropic
Hybrid - San Francisco, CA
More jobs at AnthropicRequirements
Skills
- 5-10+ years of experience building software
- High proficiency in at least one programming language such as Python, Rust, Go, or Java
- Productive use of Python
- Strong ability to learn unfamiliar domains quickly
- Excellent prioritization and decision‑making skills under ambiguity
- Preference for fast‑moving collaborative projects
- Interest in interpretability research and AI safety (research experience not required)
- Commitment to societal impacts and ethical considerations
- Ability to work closely with researchers translating research needs into engineering solutions
- Experience optimizing performance of large‑scale distributed systems
- Knowledge of language modeling fundamentals with transformers
- High‑performance LLM optimization skills including memory management, compute efficiency, and parallelism strategies
- Hands‑on experience in a mainstream ML stack such as PyTorch/CUDA on GPUs or JAX/XLA on TPUs
- Collaboration with researchers to build tooling that supports research teams
Responsibilities
- Build and maintain the specialized inference and training infrastructure that powers interpretability research, including instrumented forward/backward passes, activation extraction, and steering vector application
- Resolve scaling and efficiency bottlenecks through profiling, optimization, and close collaboration with peer infrastructure teams
- Design tools, abstractions, and platforms that enable researchers to rapidly experiment without hitting engineering barriers
- Help bring interpretability research into production safety audits with real deadlines and high reliability expectations
- Work across the stack—from model internals and accelerator‑level optimization to user‑facing research tooling
Technologies
PythonRustGoJavaPyTorchCUDAJAXXLAtransformers
See if your resume is ready for this job
See how our AI can optimize your resume and improve your chances for this role.