Latest AI JobsBrowse Jobs
Anthropic logo

Research Scientist, Interpretability

AnthropicSan Francisco, CAAI Research & Engineering
Apply on company site
Salary
Not listed
Job type
Full-time
Posted
7 Nov 2025
Status
Open · checked 7d ago

About this role

The Research Scientist, Interpretability role at Anthropic in San Francisco involves advancing mechanistic interpretability to make advanced AI systems safer and more trustworthy. This position suits individuals with a strong scientific research background who want to decode how neural network parameters map to meaningful algorithms.

Role Overview

Anthropic is seeking dedicated professionals to join its Interpretability team in San Francisco, with remote options considered for exceptional candidates. The group focuses on reverse-engineering trained models to understand how modern language models function internally. By treating neural networks like biological systems or computer programs, researchers use custom tools to solve challenges like superposition and decompose models into interpretable components.

Key Responsibilities

Create methodologies for understanding large language models through the reverse engineering of learned weights.
Design and execute experiments quickly in toy scenarios and at scale using large models.
Create and analyze new interpretability features and circuits to comprehend model mechanics.
Build necessary infrastructure to run experiments and visualize results.
Collaborate with team members to articulate findings and communicate results both internally and publicly.

Requirements

A strong background in scientific research across any field, including prior work or interest in interpretability.
Comfort with messy experimental science and inventing new methodologies.
Ability to view research and engineering as deeply connected, writing code and running experiments directly.
Clear communication skills to articulate motivations, write up results, and teach colleagues.
A minimum of a Bachelor's degree or an equivalent combination of education, training, and experience in a relevant field.

Skills and Tools

Python

Salary and Benefits

Annual salary range of $350,000 to $850,000 USD.
Competitive compensation and benefits package.
Optional equity donation matching.
Generous vacation and parental leave.
Flexible working hours.

About the Company

Anthropic is a public benefit corporation dedicated to creating reliable, interpretable, and steerable AI systems that remain safe and beneficial for society. Operating primarily out of San Francisco, the organization embraces a collaborative big-science approach, treating artificial intelligence research as an empirical science closely tied to biology and physics. The company provides visa sponsorship support through an immigration lawyer when needed and maintains an office-based hybrid policy requiring staff to be in the office at least 25% of the time.

How to Apply

Interested candidates should submit their application directly through the official Anthropic careers page.

Skills and tech stack

Questions candidates ask

Where is this job located?

The role is primarily based in the San Francisco, CA office, though exceptional candidates may be considered for remote work on a case-by-case basis.

What is the salary range for this position?

The annual base salary ranges from $350,000 to $850,000 USD.

What programming language is required?

Familiarity with Python is required for this role.

Does Anthropic offer visa sponsorship?

Yes, Anthropic sponsors visas and makes every reasonable effort alongside an immigration lawyer to secure visas for offered candidates.

What is the minimum education requirement?

A Bachelor’s degree or an equivalent combination of education, training, and/or experience in a relevant field is required.

Ready to apply?

The application is on the official Anthropic careers page.

Apply on company site