SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Eleos Health is building AI infrastructure to support behavioral healthcare delivery. You'll join the AI Platform team, a small hands-on group responsible for building the AI infrastructure, agentic workflows, and tooling that power the R&D organization.
Your mandate is to make the entire software development lifecycle faster and smarter with AI. This is not a ticket-driven role or a prompt-engineering position. You'll own a set of KPIs and be responsible for turning fuzzy goals into measurable outcomes. This means discovering problems yourself by talking to engineers across R&D, understanding where the SDLC breaks down, and building agentic workflows and platform capabilities that ship and remain reliable in production.
Key responsibilities include: owning KPIs and determining what moves them; discovering problems through direct conversation with engineers; designing, building, and shipping agentic workflows that support the R&D SDLC from prototype through production; working hands-on across agent harnesses and managing context, knowledge, and scaffolding; building feedback loops and evaluations to make agents trustworthy; extending the AI platform itself including agents, integrations, and infrastructure; prioritizing work based on impact toward team KPIs; proving concrete impact on your assigned metrics; and partnering with other engineers to keep the platform coherent.
You bring 6-8+ years of hands-on senior software engineering experience, with genuine production experience building and shipping agentic workflows and LLM-powered systems (not just prototypes). You have real experience working across agent harnesses like Claude Code, understand agent context design and scaffolding at scale, and have practical knowledge of LLMs, RAG, tool use, evaluations, and human-in-the-loop patterns. You have strong software engineering fundamentals, can work without pre-defined specs, possess product judgment to separate what's worth building, have strong discovery skills to map how work actually happens, and communicate effectively across technical and non-technical audiences.
Advantages include a cost-aware mindset regarding token usage, experience with Kubernetes/containers/CI-CD infrastructure, familiarity with MCP-based tooling or agent orchestration frameworks, and experience with observability tools like Datadog.
