TAGGED
4 articles
AllthingsPM read the 5 public sector PM job descriptions in its catalog from Anthropic, Scale AI and Sierra. All five ask for field discovery and model evaluation; two name a clearance, two name FedRAMP. Each has a mock interview built from its own text.
A Safeguards PM at Anthropic owns the detections, evals, enforcement tools and metrics that stop Claude being misused. We read all 8 safety PM job descriptions in our corpus to show what the role asks for, what it pays and how to practice for it on AllthingsPM.
AI labs hire five kinds of PM: model and research, API and platform, safety and trust, applied products, and growth. We sorted 34 live PM job descriptions from OpenAI, Anthropic and Cohere to show what each one asks for.
Anthropic's 19 PM job descriptions (September 2026, 16 live in November) pay $305,000 to $460,000 and ask for ambiguity, research partnership, safety judgment and evals far more than SQL or A/B tests. AllthingsPM lists each with a mock interview built from its own text.
Reading is the easy half.
The course grades the other half.
