San Francisco, CA

Salary
$220,000–$280,000
Posted
Jul 23, 2026
Location
San Francisco, CA
Last confirmed open
Jul 26, 2026

What this job asks for AI summary

This role focuses on investigating how generative AI systems and agents can fail or cause harm, particularly to people. Day-to-day work involves designing research programs and evaluation methodologies, leading red-teaming exercises, and translating findings into actionable guidance for product and engineering teams. It suits researchers with a background in AI risk, trust and safety, or responsible AI who can bridge rigorous research and practical governance.

Senior level · 8+ years · San Francisco-Oakland-Hayward, CA · Master's required · Full-time

Must have (3)
AI/ML systems · 5+ yrsqualitative research methodsquantitative research methods
Nice to have (4)
generative AIAI red teamingadversarial testingdata science

We read this from the posting text with AI. Skim the description below before ruling yourself out.

How this req sits in the market our data

What won't set you apart
AI/ML systems80%

Most people in this occupation already list these. Still required — just not what gets you shortlisted.

What the occupation pays Median $173,851 (middle half $133,298–$217,653).

Estimated from BLS employment for this occupation and area, per-skill prevalence across our listing corpus, and published wage benchmarks — as of Jul 28, 2026. It is a model, not a headcount.

Why we read it this way (6)

This role is highly interdisciplinary and does not map cleanly to a single SOC code. The primary day-to-day work is AI risk research — investigating failure modes, red-teaming, and designing evaluation methodologies for generative AI systems — which sits closest to 15-2051 (Data Scientists) given the research, mixed-methods, and analytical orientation. 15-1299 (Computer Occupations, All Other) is a plausible runner-up given the responsible AI / AI safety focus that doesn't fit neatly into any standard occupation.

The degree requirement states 'master's or other advanced degree … or equivalent industry experience.' Because equivalent experience is explicitly accepted, the degree requirement is set to Masters rather than None — the posting does gate on an advanced degree OR equivalent, but the 'or equivalent' clause means it is not a hard academic gate. This is a borderline call; the Masters flag reflects the posting's clear preference for an advanced degree.

The 8+ years of experience requirement is stated at the role level (research involving technology, human behavior, risk analysis, etc.) and is captured in the overall years minimum. The 5+ years with AI/ML systems is technology-specific and is captured on that skill's years demanded.

No compensation figures are disclosed in the posting.

The role is explicitly on-site in San Francisco; remote is set to false.

Ignored 1 non-technology phrase(s) as skills (responsibilities/concepts, not named tools): risk analysis.

Read the full posting

The employer publishes the full description on their own site — read it there ↗. Or sign in to read it here — it's free, and it also lets you track this application.

Apply

Apply on employer site ↗