Agentic AI Engineer

This role sits at the core of Zoommate's agent mission, building the harness, including runtime, orchestration, tool-use, memory, and evaluation layers, that every downstream agent product depends on. You will shape the shared infrastructure that multiplies leverage across all of Zoom's product teams, working hands-on from design to production. Joining a lean, high-engineering-density team means short decision paths, end-to-end ownership, and the opportunity to blur the lines between agent infrastructure, applied research, and product integration.

This is a hybrid position that requires on-site presence 4 days per week at our Seattle, WA office. Remote arrangements are not available for this role.

This role sits at the core of Zoommate's agent mission, building the harness, including runtime, orchestration, tool-use, memory, and evaluation layers, that every downstream agent product depends on. You will shape the shared infrastructure that multiplies leverage across all of Zoom's product teams, working hands-on from design to production. Joining a lean, high-engineering-density team means short decision paths, end-to-end ownership, and the opportunity to blur the lines between agent infrastructure, applied research, and product integration.

This is a hybrid position that requires on-site presence 4 days per week at our Seattle, WA office. Remote arrangements are not available for this role.

The Zoommate team is a lean, high-performance group that owns agent outcomes end-to-end, from the first design doc to production across Zoom's product suite. We operate with high engineering density and short decision paths, blurring the lines between agent infrastructure, applied research, and product integration. We hire for depth across backgrounds, whether backend engineering, research engineering, or applied science, and we ship bigger things with a smaller, sharper crew.

• Design, build, and evolve the Zoommate agent harness, including orchestration, tool invocation, sandboxing, memory and state management, context management, and multi-step reasoning loops that other teams build on top of.
• Architect the underlying agent framework and API/SDK surface so Zoom product teams can embed Zoommate Agent capabilities with minimal integration cost.
• Build and integrate tool-use and interoperability layers, including MCP-style connectors and agent-to-agent patterns, so agents can act across Zoom's ecosystem and third-party apps.
• Design and run evaluations and benchmarks for agent quality across reasoning, tool-use success, latency, and cost, and use findings to drive continuous iteration.
• Own the full lifecycle of services you build, including design docs, code review, deployment, monitoring, and on-call, with production-grade reliability at Zoom scale.
• Partner closely with product teams to turn real integration pain points into better shared infrastructure rather than one-off fixes.
• Integrate with Zoom's engineering platforms, leveraging internal tooling to accelerate delivery.
• Apply AI coding agents as first-class teammates in your daily workflow and bring that intuition into how you design tools for agents.

• 5+ years of industry experience building production software and/or ML systems, with hands-on experience delivering agentic systems using LLMs (tool use, planning and reasoning, memory, orchestration, or multi-agent coordination).
• Strong programming fundamentals across data structures, algorithms, concurrency, and system design; fluency in Python and comfort with at least one of Java, Go, or Type Script.
• 2+ years of API/SDK development experience (REST, gRPC, or similar) with demonstrated ability to design infrastructure…