Mechanistic Interpretability

Acceler8 Talent; San Francisco (CA); On-site; USD 140,000 - 190,000; Full time

Job summary

Acceler8 Talent is seeking a Research Engineer focused on Interpretability Systems in the San Francisco area. The role centers on understanding what happens inside large language models, not production ML, MLOps, or large-scale training infra.

You will build experimental systems, develop tooling, and run fast, greenfield research cycles to test ideas and push forward interpretability, alignment, and RL research. This is a hands-on, lab-focused role for curious engineers.

Qualifications

• Strong software engineering fundamentals.
• Experience with experimental ML / research systems.
• Comfort working close to model internals.
• Interest in interpretability, alignment, RL, or mechanistic understanding.
• PhD helpful, not required.

Responsibilities

• Research and develop interpretability tooling and experiments.
• Conduct activation tracing and mechanistic analysis of language models.
• Create benchmarks and tools to advance interpretability research.

Skills

Software engineering fundamentals

Experimental ML / research systems

Understanding model internals

Interest in interpretability / RL

PhD helpful

Job description

Research Engineer – Interpretability Systems

An AI research lab working at the frontier of interpretability, alignment, and reinforcement learning is hiring Research Engineers focused on understanding what's happening inside large language models

This role is for engineers who want to build the experimental systems that make interpretability research possible - not production ML, MLOps, or large-scale training infra

You'll work on:

• Activation tracing & mechanistic analysis
• Activation-level steering beyond prompting and fine-tuning
• New benchmarks for model consistency and robustness

The work is fast, experimental, and greenfield: build custom tooling, test research ideas, get results, move on.

Ideal background:

• Strong software engineering fundamentals
• Experience with experimental ML / research systems
• Comfort working close to model internals
• Interest in interpretability, alignment, RL, or mechanistic understanding
• PhD helpful, not required

This is not a role for scaling pipelines or maintaining production systems

It's for people who enjoy ambiguous problems, fast research cycles, and building new tools from first principles