Senior GenAI Safety & Evaluation Engineer

Google; United States; On-site; USD 159,000 - 230,000; Full time

Benefits offered by this job

Health insurance

Dental insurance

Vision insurance

Life insurance

Disability insurance

401(k) match

Paid time off

Sick time

Job summary

Google is seeking a Senior Engineering Analyst in Trust and Safety SAGE to lead automated safety evaluation platforms powering next-gen generative AI products. You will transition from bespoke scripts to high-throughput autoraters, enabling real-time launch decisions and offline testing with robust safety guardrails.

You will tackle challenges in health, civics, and AI ecosystem safety, analyzing risks and influencing cross-functional partners while maintaining high standards of integrity and

Qualifications

• Bachelor's degree or equivalent practical experience.
• 5 years of experience in data analysis, identifying trends, statistics, and insights.
• 5 years of experience in the anti-abuse domain focusing on signal analysis, model development, data analysis, and cybersecurity.
• Experience programming in SQL, Python, or Golang.

Responsibilities

• Establish metrics and benchmark Generative-AI issues; develop datasets and run evals for engineers.
• Drive project delivery with technical judgment and manage interdependent projects of moderate to high complexity.
• Improve tools via data analysis and stakeholder presentations; enhance workflows through automation opportunities.
• Design, develop, and scale production-grade LLM-as-a-judge autoraters and evaluation systems for real-time decisions.
• Partner with Product, Search Core, and GDM engineering teams to align on technical roadmaps and risk registers.

Skills

Data analysis

Anti-abuse domain

Programming

Mentoring

Education

Bachelor's degree or equivalent

Tools

SQL

Python

Golang

Job description

Google is seeking a Senior Engineering Analyst in Trust and Safety SAGE to lead automated safety evaluation platforms powering next-gen generative AI products. You will transition from bespoke scripts to high-throughput autoraters, enabling real-time launch decisions and offline testing with robust safety guardrails.

You will tackle challenges in health, civics, and AI ecosystem safety, analyzing risks and influencing cross-functional partners while maintaining high standards of integrity and