Scale AI Has 6 Open AI Product Manager Roles. OpenAI Has Zero.
OpenAI has 97 open AI/ML roles in HireIndex’s August 24, 2026 index — more than any company tracked. It has zero open AI Product Manager roles.
Neither does Cohere (19 total roles). Nor Perplexity (18). Nor Figure AI (17). Nor Writer (25). Nor Coreweave (15). The list of major AI companies with no AI PM openings runs longer than the list of companies that have them.
Scale AI, by contrast, has 6. That’s 10.5% of their 57-role roster. Plaid has 2 out of 10 — the highest ratio among tracked companies at 20%. Royal Bank of Canada has 2 out of 13.
The question that number raises: what does it mean that the company building the most advanced AI models in the world doesn’t have a single AI PM role open, while a financial data API company does?
How many AI PM jobs are open right now?
34, as of the August 24, 2026 snapshot. HireIndex tracks 1,385 AI/ML roles across 498 companies; AI Product Manager accounts for 2.5% of that total.
For scale: Machine Learning Engineer roles number 274. Data Scientist: 165. AI Software Engineer: 114. MLOps Engineer: 44. LLM Engineer: 29. AI Product Manager slots in below all of them at 34, just ahead of Data Engineer (9).
It’s a small category. But the distribution of who holds those 34 roles tells a story about where AI product work actually lives.
Which companies are hiring AI PMs?
| Company | AI PM roles | Total AI roles | AI PM share |
|---|---|---|---|
| Scale AI | 6 | 57 | 10.5% |
| Databricks | 3 | 71 | 4.2% |
| Cresta | 2 | 21 | 9.5% |
| PointClickCare | 2 | 21 | 9.5% |
| Royal Bank of Canada | 2 | 13 | 15.4% |
| Plaid | 2 | 10 | 20.0% |
| 1 | 38 | 2.6% | |
| J.P. Morgan | 1 | 11 | 9.1% |
| 1 | — | — | |
| Hyatt | 1 | — | — |
| U.S. Bank | 1 | — | — |
Scale AI’s six roles break down as: four senior-level (“Senior AI Product Manager, Code,” “Senior AI Product Manager, Cybersecurity,” “Senior AI Product Manager, Finance Agents,” “Senior AI Product Manager, Leaderboard”) and two mid-level (“AI Product Manager, Coding/Multimodal” and “Product Manager of AI Applications, Global Public Sector”). The vertical specialization is the tell — these aren’t generalist PM roles. Each one maps to a specific customer industry.
Databricks’s three PM roles sit at Staff or above: “Staff Product Manager, Agentic AI Applications,” “Staff Product Manager, AI Platform,” and “Sr. Product Manager, Databricks AI.” All three are remote.
Plaid’s two PM roles (“Staff Product Manager - AI Foundations” and “Staff Product Manager - Internal AI”) are both internal-facing — the PM layer managing Plaid’s own AI build-out, not customer-facing product work. That’s a different job from Scale AI’s vertical PMs, but the same title.
Why do the frontier labs have no AI PM roles?
The companies with the most AI/ML open roles — OpenAI, Mercor, Databricks, Scale AI, Reddit — divide into two camps on this.
OpenAI, Cohere, Perplexity, Figure AI, Coreweave, Writer: zero AI PMs. Their product surface is code. Developers and enterprises consume it through APIs or direct contracts. The product direction comes from founders and researchers; there’s no customer-requirements translation layer that a PM typically owns.
Scale AI, Databricks, Cresta, Plaid: meaningful AI PM representation. These companies sit in a different position: they build tools that other companies use to apply AI. Scale AI’s platform helps enterprises run AI evaluation and training. Databricks gives companies a place to run data pipelines and AI workloads. Cresta runs AI for contact centers. In each case, the PM’s job is to figure out what a specific set of customers — finance, cybersecurity, public sector — actually needs from the product.
The pattern holds at the infrastructure layer too. Coreweave (15 roles) and Deepgram (15 roles) have zero AI PM openings; their products go to technical buyers who evaluate them on performance specs, not feature roadmaps shaped by a PM.
What does the seniority breakdown look like?
There are no junior or entry-level AI PM roles in the August 24 index. None.
Breakdown: mid-level (15 roles), senior (12), senior-heavy (7). The floor is mid. That’s consistent with how the role works — you need enough product experience to manage requirements across stakeholders and enough AI literacy to make real tradeoffs on implementation. A bootcamp graduate or associate PM doesn’t have either.
The implication for candidates: this is a category you arrive at, not one you start in. The typical path in the index involves prior PM experience plus a meaningful AI project or domain — either PM’ing a data/ML product, working in a technical domain the role covers (finance, healthcare, cybersecurity), or internal AI adoption work.
Where are AI PM roles?
73.5% remote — 25 of 34 — compared to 67.7% remote across all AI/ML roles in the index.
Non-remote locations: London has 2 (J.P. Morgan, Depop), San Francisco has 2 (Plaid), New York has 2 (Google, Red Ventures), Berlin has 1 (voize), Sydney has 1 (Onset).
Toronto has 72 open AI/ML roles in the same snapshot and zero AI PM roles. Seattle and Boston also show zero. The category hasn’t built geographic density outside London and San Francisco, and even there the numbers are thin.
What’s the non-AI-company story?
Thirteen of the 34 roles are at companies that exist for something other than AI. Hyatt runs hotels. U.S. Bank runs checking accounts. Berkadia does commercial real estate. Depop resells fashion. Red Ventures runs media brands. Motorola Solutions builds radios and cameras for law enforcement.
Each of them has an open AI PM role.
The job at Hyatt is “Data & AI Product Manager.” At U.S. Bank it’s “AI Product Manager,” classified mid-level. At Berkadia it’s “Product Manager - AI,” also mid. At Depop it’s “Senior Product Manager - Ranking (ML),” where ML means the algorithm that determines what a buyer sees when they open the app.
These aren’t AI companies trying to build frontier models. They’re businesses that run on existing software and are now integrating AI into it — pricing algorithms, recommendation systems, underwriting models, fraud detection. Someone has to own that integration as a product function and translate what the engineering team can do into something the business unit will actually use. That’s the AI PM role at these companies, and it’s a different job from the Scale AI version.
Plaid’s “Staff PM - Internal AI” is a third variant: a PM who owns the company’s own internal AI adoption. As AI coding tools and AI operations tools proliferate, companies are finding that someone needs to own the internal rollout as a product — not an IT project, not a one-time deployment, but an ongoing function. That role is starting to appear in job listings under the AI PM title.
The August 24 snapshot has 34 of them.