You and the process

With a person
Anna, in-house recruiter
9 years hiring engineers
30 minutes with a real recruiter
They read your CV with you, on a call, and say where the offers are being lost.
Didn’t find what you were looking for? Tell us what to build
Anthropic

Staff+ Software Engineer, Safeguards Evals

  • Hybrid
  • 6+ years

Salary

$320,000 - 485,000/ year

AI summary

For members

The whole posting in a few lines. Sign up to read it here and on every role you open.

Sign Up to Read

Description

The company builds AI systems meant to be safe and useful for users and society. The team is growing and includes researchers, engineers, policy experts and business leaders.

The team answers two questions with evaluations: whether a model is safe, and whether the systems that catch misuse actually catch it. Evaluations cover misuse, prompt injection and user well-being, and feed training and deployment decisions. The company also uses AI to examine real Claude traffic for bad actors and new threats that lead to enforcement.

The role builds the methods and infrastructure behind these evaluations. You will work between applied ML research and engineering, designing experiments, building datasets that reflect real abuse, and moving the resulting methods into the pipelines that gate training, agent changes and launches.

Responsibilities

  • Run experiments that make evaluations more accurate. Build ways to generate representative test data, simulate realistic users and check grading accuracy. Study how multi-turn conversations, tools, long context and user diversity affect safety behavior. Find gaps in evaluation coverage and update evals so they stay challenging and informative as capabilities grow.
  • Build and own the evaluation harness for an agentic investigation system. Define metrics, test cases and grading for a complex, long-horizon agent. Measure end-to-end performance, including detection precision and recall, investigation quality and robustness, and push improvements on the hardest harm areas. Build RL environments that improve Claude's safety investigation skills.
  • Create high-quality eval datasets of real-world misuse in areas such as cyber attacks, bio weapons and influence operations, using real traffic patterns and synthetic generation. Work with Policy and Enforcement to turn observed harm patterns into measurable evaluations.
  • Move successful research into evaluation, regression and release pipelines that run during training, on every agent change, prompt update and model upgrade, and after launch. Build tools that let policy experts write, run and refine evaluations without engineering help. Share findings with research and training teams to improve models upstream.

Requirements

  • Experience building and maintaining data pipelines.
  • Experience working with LLMs, with a working understanding of their capabilities and failure modes, especially agentic systems that use tools and multi-step reasoning.
  • Strong data analysis skills, with the ability to draw reliable conclusions from large datasets.
  • Ability to move between research prototypes and production-quality code.
  • Ability to turn ambiguous problems into concrete, testable experiments.
  • You care deeply about AI safety and want your work to matter.
  • Preferred: 8+ years of industry software or ML engineering experience.
  • Preferred: experience building or contributing to LLM or agent evaluation frameworks, benchmarks or automated grading systems.
  • Preferred: extensive experience in trust and safety, content moderation or abuse detection.
  • Preferred: experience in red teaming, adversarial testing or jailbreak research on AI systems.
  • Preferred: experience with synthetic data generation or data augmentation.
  • Preferred: experience with distributed systems or large-scale data processing.
  • Preferred: experience with prompt engineering or building LLM-powered applications.
  • Minimum education: a Bachelor's degree or an equivalent mix of education, training and experience, in a field relevant to the role as shown by coursework, training or professional work.
  • Years of experience required follow the internal job level for the position.

Conditions

Annual base salary range: $320,000 to $485,000 USD.

Location: hybrid. Staff are expected in an office at least 25% of the time, and some roles may require more.

Visa sponsorship is available, but not for every role or candidate. If an offer is made, the company will make every reasonable effort to obtain a visa and works with an immigration lawyer.

The company encourages applications even when you do not meet every qualification, and welcomes diverse perspectives.

Benefits

  • Competitive compensation and benefits.
  • Optional equity donation matching.
  • Generous vacation and parental leave.
  • Flexible working hours.
  • An office space for collaboration with colleagues.

Where you’d work

Part of the week in the office

About the company

Anthropic

  • Industry: AI

Offices in New York City, United States, San Francisco, United States, Seattle, United States

Also hiring in Paris, France, London, United Kingdom, Boston, United States and 5 more places

12 of their 119 open roles are remote

Your chances

Worth a look before you spend an evening tailoring a CV for it.

  • 22 checks run
  • 3 red flags

Still hiring?

14 checks

No red flags

How crowded?

8 checks

3 red flags

Fits Me

How well does this role fit you?

Answer a few questions or drop your CV, and every role gets a fit score with the reasons, this one first.

  • Your field
  • Level
  • Stack
  • Work model
  • Salary floor
  • Must-haves
Details13 facts · Role, Location, Compensation, Company
Tech stack
  • LLMs
Seniority
Staff
Equity
Equity offered
Industry
AI
Specialty
Backend
Region
United States
Show 7 more factsShow less

Role

Category
Development
Specialty
Backend
Seniority
Staff
Experience
8+ years
Tech stack
  • LLMs

Location

Work model
Hybrid
Region
United States
Offices
  • New York City, United States
  • San Francisco, United States
  • Seattle, United States
Visa sponsorship
Sponsored

Compensation

Salary
$320,000 - 485,000 / year
Pay period
Annual
Equity
Equity offered

Company

Industry
AI

Something wrong with this vacancy?

Similar vacancies

Share this vacancy

What's wrong with it?

The employer never sees who reported.

Reason