Senior Product Engineer

Talkie builds AI Agents for healthcare. These agents handle patient-clinic communication end-to-end over voice calls, web chat, and SMS, providing 24/7 access to care. The company serves primary care and specialty practices, and hospitals in the US and Poland, handling close to a million real patient conversations monthly.

Must have:

• Product ownership: experience shipping and owning products or major features end-to-end.
• Strong architectural experience: building maintainable and adaptable software, with clear trade-offs.
• Fullstack TypeScript experience: solid hands-on with Node.js, React, and MongoDB or another database.
• Cloud and infrastructure as code experience: hands-on with AWS or another major cloud provider and IaC tools like Terraform and Pulumi.
• AI-native development: proficient use of coding agents as a main way of working.
• LLM integration in production: built LLM-powered features with evals, managing latency and cost trade-offs.
• Healthcare compliance awareness: knowledge of SOC2 and HIPAA practical implications.
• Product sense for UI: ability to design clean UI and user interactions without detailed mockups.
• English proficiency at C1 level or above.

Nice to have:

• Telecom experience with SIP, call routing, or VoIP.
• Experience with EHR or healthcare integrations involving clinical systems or regulated domains.

• Own product features end-to-end: research problems, refine requirements with the CPO, propose user interaction, design UI, build, ship, and monitor.
• Build across the stack: write backend services and APIs in TypeScript and Node.js, frontend in React, and manage AWS infrastructure as code.
• Integrate LLMs into the product: design agent interactions with models and tools, build evals for quality, latency, and cost trade-offs.
• Integrate with outside systems including SIP telephony, call routing, EHR systems, and LLM providers.
• Shape technical and architectural decisions through design discussions, proposals, and maintaining platform maintainability and real-time call efficiency.
• Mentor other engineers by reviewing designs and code, pairing on complex problems, and helping them grow.
• Turn open questions into actionable plans by building prototypes, measuring results, and proposing trade-offs.
• Keep production healthy by monitoring releases, reducing alert noise, and fixing performance bottlenecks.

AI-native development:

• Orchestrate multiple coding agents in parallel for various tasks such as implementation, code review, log investigation, and research.
• Improve the codebase for agents by ensuring clear module boundaries, good agent instructions, and fast feedback loops.
• Build reusable skills, commands, and workflows for agents to operate correctly.
• Share best practices and teach the team effective agent workflows, recognizing when to distrust agent output.