Forward Deployed Research Scientist

Goodfire•San Francisco, CA

28d•$220,000 - $315,000•Onsite

About The Position

Goodfire is a research company using interpretability to understand, learn from, and design AI systems. Our mission is to build the next generation of safe and powerful AI—not by scaling alone, but by understanding the intelligence we're building. Treating models as black boxes is an unnecessary handicap—we have access to the structures inside them, and understanding those structures lets us steer what models learn, make them safer and more useful, and extract the vast knowledge they contain. Our goal is to make AI that can be understood, debugged, and shaped like software. Goodfire is a public benefit corporation headquartered in San Francisco with a team of the world’s top interpretability researchers and engineers from organizations like OpenAI and DeepMind. We're backed by over $200M from B Capital, Menlo Ventures, Lightspeed, Eric Schmidt, and others. We’re looking for a Forward Deployed Research Scientist to join our team to work directly with customers to apply our interpretability platform to solve novel problems for our customers. You’ll work closely with a small, mission-driven team of scientists and engineers to conduct novel research, build practical tools, and push the field forward.

Requirements

MS/PhD or equivalent experience in CS, with a strong ML background and experience in interpretability.
Deep familiarity with large models and a passion for understanding how they work.
Fluency in Python and ML frameworks such as PyTorch.
Strong writing and communication skills for explaining complex ideas.
Drive to move quickly and take ownership.

Nice To Haves

Experience leading research in a forward deployed environment and in service of a customer’s requirements.
Experience in startup or fast-paced lab environments.

Responsibilities

Lead research with our partners to interpret advanced models.
Own research delivery on high-stakes projects with customers and do whatever it takes to make delivery successful, including: problem and hypothesis definition, data sourcing, tool building, iteration, and implementation.
Translate research into tools for real-world applications.
Work directly with customers to conduct original research in interpretability.
Prototype techniques to visualize and manipulate internal model structures.
Collaborate with engineering to turn research into production-ready tools.
Drive partnership success by collaborating closely with our customers and partners.
Help define and evolve our research direction.