Where the hiring is
All 3,027 tracked AI roles by category. Agent engineering dominates the market; governance and security are still small — but they're the fastest-growing edge.
Two markets, side by side
Same role taxonomy, different shape and different employers.
Forbes AI 50 + CB Insights AI 100
1,822 AI roles · 134 companies
- Agent engineering 64%
- AI/ML engineering 20%
- Agent ops & infra 6%
- AI product 4%
- Field & solutions 4%
Top hirers: Salesforce (137), Anthropic (109), Databricks (108), OpenAI (105)
Curated Fortune 500 deployers
1,205 AI roles · 69 companies
- Agent engineering 59%
- AI/ML engineering 29%
- Agent ops & infra 5%
- AI product 4%
- AI security 1%
Top hirers: Citi (99), Capital One (78), Netflix (65), Autodesk (65)
Where they differ
Both lean on agent engineering — the divergence is everywhere else.
Read maturity from the hiring
Each company placed by the most advanced role it posts — frontier and enterprise shown separately.
Anthropic, Databricks, OpenAI · Citi, Capital One, Netflix
Salesforce, Palantir, Okta · Disney, Visa, Cigna
Layer Health, World Labs, EliseAI · AeroVironment, Wolfspeed, Regions Bank
Emerging agentic titles
New role concepts only one company is naming so far — matched on the role itself, so the same job in a new location or grade isn't counted as new.
How the index is built
Two cohorts, classified the same way on the full job description: frontier (Forbes AI 50 + CB Insights AI 100) and a curated panel of Fortune 500 enterprises. A role counts as agentic only on a hard signal — a named framework, a protocol like MCP, or a multi-agent architecture — or two distinct softer signals; a generic "agentic AI" mention is not enough. We link out to every careers page and host no listings. Dataset CC BY 4.0 — JSON · CSV · full company tracker · the index in depth.