Remote job
Feltsense - AI Engineer
Job details
About this role
Role overview Own the architecture and hands-on development of an autonomous agent platform that can take products from opportunity discovery through launch. This technical leadership role combines production engineering with decisions about system design, agent orchestration, reliability, and engineering practices.
Responsibilities - Set technical direction and make architecture decisions across the platform. - Build multi-agent systems and orchestration infrastructure for end-to-end product workflows. - Develop agent improvement mechanisms, including feedback loops, evaluation, and memory. - Create and maintain backend services that use large language model APIs. - Ship and support production code, improving agent behavior through prompt design, structured outputs, retries, and fallback handling. - Turn complex problems into engineering milestones, establish practices for code review, testing, observability, and releases, and support the team while contributing directly.
Requirements - At least four years of professional software engineering experience. - Previous technical leadership at an early-stage company or in a comparable engineering leadership role. - Hands-on experience building and shipping multi-agent systems or orchestration frameworks in production. - Experience developing self-improving or self-evolving agent systems. - Strong backend skills, including designing and maintaining APIs, services, and data flows. - Experience with LLM-based systems, including prompt engineering, model configuration, structured outputs, embeddings, retrieval, and evaluation. - Regular use of AI coding agents, with the ability to delegate tasks, review generated work, and verify results.
Nice to have - TypeScript and Node.js experience, or familiarity with agent SDKs and custom runtimes. - Experience with PostgreSQL, Supabase, pgvector, background job systems, Zod, Vitest, or Cloud Run. - Experience growing distributed engineering teams, or an interest in AI evaluation, safety, and observability.