ABOUT ODE WITH ANTHROPIC
How do you turn a decades-old dataset into an industry-leading medical coding agent? Teach an AI receptionist to book appointments in Spanish? Automatically generate working data connectors from API docs? Build agents that write, test, and deploy their own software packages?
Anthropic is defining frontier AI. Ode is the team bringing frontier AI into practice.
We are a team of experienced builders who care deeply about getting complex AI systems into production, with strong conviction about what makes them succeed. As Anthropic’s services venture, we tackle the hardest, most consequential AI problems for companies with real stakes, working closely with their Applied AI and Product teams to turn frontier technology into real world impact.
We're backed by Anthropic, Blackstone, Hellman & Friedman, Goldman Sachs, General Atlantic, Leonard Green & Partners, Apollo Global Management, GIC, and Sequoia Capital.
We're headquartered in SF with offices in NYC, Raleigh-Durham, and Dubai.
THE ROLE
Our customers come to us with ambitious, often fuzzy goals. They want new AI capabilities (agents, workflows, skills) built on messy proprietary data that hold up at scale, and they often can't yet say what success looks like.
Your job is to turn those goals into AI-powered solutions that work in the real world, and to own the journey from project kickoff to live system to a tangible business outcome, not just a demo.
This is a technical, hands-on product role. You work directly in the AI systems you build with our engineering team: you prototype early, work with the data, run evals, and help hill-climb on quality. You make product-level technical calls yourself and act as a thought partner to the lead engineer on architecture, weighing short-term velocity against long-term extensibility.
You also work directly with clients. You build trust with their stakeholders and present the work to their leaders.
This is a forward-deployed, full-stack product role:
- Part product manager: you decide what we build and why.
- Part builder: you prototype fast and get hands-on in the AI systems alongside our engineering team.
- Part delivery manager: you keep execution on track and keep client stakeholders aligned on progress, risks, and tradeoffs.
- Part "do whatever it takes" operator: you jump into the gaps to keep things moving.
WHAT YOU’LL DO
- Shape what we build and why. Translate ambiguous business problems into crisp problem statements, user journeys, and success metrics, and work with customers to define what "success" actually means.
- Turn messy problems into shipped solutions. Translate ambiguous needs into clear requirements and PRDs, and run planning with engineers. Bring focus to define what good looks like for an AI system, then the drive to get it there.
- Prototype to discover and derisk. Build working prototypes early to learn what users need and to surface the hardest technical risks before the engineering team commits to a build.
- Build the ground truth. Work with client experts to assemble and label the ground-truth data that defines correct behavior for an AI system. Check it for coverage, consistency, and gaps, and plan how it gets maintained and extended.
- Hill-climb quality with the engineering team. Work with engineers on prompts, context design, and evals for the AI systems you build together. Run offline evals against ground truth during development and online evals on live traffic after launch. Read the failures, propose the fix, and repeat until we hit production quality.
- Shape technical decisions. Make product-level calls on model choice, retrieval and agent design, context management, and build vs. buy, and own the tradeoffs those calls carry for the project.
- Drive execution across multiple AI products. Run the day-to-day rhythm including planning, check-ins, risk tracking, and coordinating between clients, engineers, designers, and Ode teammates to keep work unblocked and on time.
- Drive…