Technical Product Manager, Labeling Tools

Centaur AI is a modern data labeling solution driving AI innovation across the healthcare industry. The company is seeking a Technical Product Manager to own the full labeling tools surface, set its roadmap, run engineering workstreams, prototype annotation workflows, and ship product features and fixes. The role focuses on expanding support for medical imaging, unstructured text, multi-modal reports, and longitudinal medical data.

Responsibilities:

• Set direction for the full labeling tools surface: our medical imaging annotation tool, web labeling, and the tooling around both
• Decide which task types and annotation capabilities we support next, what they unlock for customers, and what we say no to
• Partner with commercial and operations teams to translate customer needs into tooling capability, rather than one-off fixes
• Lead sprint planning and execution for labeling tools, including backlog grooming and release planning
• Triage incoming issues across the surface, including urgent customer-driven work, and route or resolve them
• Keep engineering unblocked by resolving ambiguity early and holding scope
• Prototype new annotation workflows to test ideas before committing engineering time
• Ship smaller features and fixes directly
• Close the loop between what we learn from labelers and customers and what actually changes in the product
• Use coding agents, LLM APIs, and MCP-based workflows to move faster, and spot manual workflows across labeling and tooling that an agent should be handling instead

Requirements:

• Engineering degree or equivalent hands-on background
• 3+ years of experience owning a technical product surface, whether as a PM, TPM, or an engineer who moved toward product
• Track record running sprints and triage for an engineering team
• Fluency with AI tools in your daily work, with specific examples of what you have built with them
• Comfort being the most technical person in a product conversation and the most product-minded person in a technical one
• Bias toward talking to users. Our labelers are diverse and sophisticated, and our tasks can be very complex
• Healthcare, medical imaging, or clinical workflow exposure
• Experience with medical imaging viewers or annotation tooling
• Familiarity with data labeling operations or crowdsourced quality control
• Experience shipping agent-based or LLM-powered features in production