Staff Engineer

Role overview Provide hands-on technical leadership for an AI-powered investigation platform used in high-stakes cybersecurity response and recovery work. You’ll build agent capabilities, measure the quality and evidentiary support of their findings, and strengthen the services that support the platform as it grows. The position calls for sound architectural judgment, a willingness to work through ambiguity, and the ability to raise engineering quality while remaining closely involved in implementation.

Responsibilities - Build investigation tools and integrations with security data sources; improve how agents manage context, retrieve evidence, and balance latency with cost. - Develop evaluation systems that test whether agents produce correct, well-sourced findings, and use results to guide model, prompt, and tool changes. - Evolve service boundaries and failure handling as the platform adds capabilities. - Improve practices for API governance, testing, and observability through design and code reviews and reusable engineering patterns. - Partner with product management to turn investigation needs into technical specifications. - Design and maintain event-driven ingestion pipelines that move structured data from object storage into analytical systems.

Requirements - Track record designing and shipping distributed systems that have operated in production over time; experience typically gained in a staff, principal, or lead engineering role. - Production experience taking AI agents beyond prototypes across multiple model providers, including tool-calling loops, regression-catching evaluation harnesses, and safeguards against unsupported outputs. - Deep expertise in Go or Python and strong proficiency in the other, including runtime behavior, concurrency, and memory management. - Experience with relational systems of record, columnar analytical stores, and vector search, plus zero-downtime schema migrations. - Experience designing schema-first, contract-driven APIs with generated clients and validation, including streaming interfaces. - Experience building durable workflows and event-streaming systems, with idempotency, retries, and recovery from partial failures built into the design. - Daily use of coding agents, with the ability to review and verify generated code.

Nice to have - Cloud architecture, infrastructure as code, and container orchestration experience, including approaches to isolating data by customer or case. - Experience building security products or familiarity with digital forensics, incident response, or security telemetry.