Machine Learning Infrastructure Engineer, Safeguards Research at Anthropic
New York, NY
$350,000–$500,000
Jul 20, 2026
New York, NY
Jul 22, 2026
What this job asks for AI summary
This role focuses on building and maintaining the ML infrastructure that underpins AI misuse detection research — covering data pipelines, training and evaluation workflows, and tooling that researchers use directly. The work sits at the boundary of research and production, requiring someone who can keep the experimentation loop fast while ensuring results remain correct and reliable as models evolve. It suits engineers with a background in large-scale distributed or data-intensive systems who are keen to develop deeper machine learning expertise.
Senior level · San Francisco-Oakland-Berkeley, CA · Full-time
Posted 4 times — it's one opening, so apply once.
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
Roughly 1,950 people in the San Francisco-Oakland-Berkeley, CA area plausibly meet what this posting asks for (software developers). range 1,450–2,550
Applicant volume Moderate — A normal amount of company. The rare requirements below are what will separate a shortlisted application from the rest.
Most people in this occupation already list these. Still required — just not what gets you shortlisted.
What the occupation pays Median $190,744 (middle half $167,095–$224,501).
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)
The role sits at the intersection of ML research infrastructure and data/distributed systems engineering — the primary day-to-day work is building pipelines, tooling, and abstractions (software development), but a meaningful portion involves large-scale data pipeline architecture, making 15-1243 a plausible runner-up.
No explicit location is stated in the posting; Anthropic is headquartered in San Francisco, CA, so that CBSA is inferred. The posting does not confirm remote eligibility.
No compensation range is disclosed.
No minimum years of experience is stated at the role level.
'Distributed systems' and 'data pipelines' are listed under Minimum Qualifications with firm language ('Experience building and operating data-intensive or distributed systems in production') and are treated as hard gates, even though they name capabilities rather than specific named tools. All preferred qualifications (ML systems, transformers, GPU programming, inference optimization, experiment tracking, evaluation harnesses, probes/interpretability) appear under the explicit 'Preferred qualifications' heading.
Ignored 1 non-technology phrase(s) as skills (responsibilities/concepts, not named tools): inference optimization.
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.