Staff Platform Engineer - AI

Staff Platform Engineer - AI

We're looking for a Staff Platform Engineer to be a senior technical voice on a brand-new team building the platform that helps Datavant teams run AI agents safely and reliably. You bring the seniority that comes from designing and operating platforms others have depended on, and the discipline that comes from doing it in regulated environments where audit, identity, and policy are real, not theoretical. You write code daily, prefer shipping the smallest thing that solves the problem to designing the perfect solution that doesn't ship, and build guardrails rather than gates.

You partner widely without claiming territory that isn't yours. You believe a platform is only adopted when it makes the secure, observable, governed path the easiest path. You mentor when you're the most senior person in the room, ask questions when you're not, and treat decisions as artifacts that get written down so reviewers can reconstruct the reasoning months later.

We practice compound engineering: each platform component we ship is also an investment in how the next one gets built. We work in tight loops (build, evaluate, refine, codify) and treat the patterns we learn as artifacts encoded into tooling, templates, and reusable substrate rather than tribal knowledge. As a Staff Engineer, you set this tempo as much as you participate in it.

What You Will Do

• Help shape the AI Platform's core architecture alongside the VP of AI Platform Engineering and a small group of senior engineers.
• Establish the platform's technical contracts: how agents register, how they prove identity, how their actions get attributed, audited, and constrained at runtime.
• Set the technical bar for the team. Mentor two to three engineers, review the work that ships, and author the design documents future hires will read to understand how things are built.
• Design how the AI Platform interlocks with the rest of engineering: the platform teams whose substrate it builds on, and the product teams whose work it serves.
• Carry the pager. Operate what you build.

What We're Looking For

• Ten or more years as a software engineer building production platforms, with at least three at Staff IC level or equivalent technical-lead seniority.
• Deep experience designing and operating cloud-native, Kubernetes-based platforms in production. AWS experience is a plus.
• Hands-on fluency with at least one mainstream LLM application stack (LangGraph, Bedrock AgentCore, or equivalent), or strong adjacent platform experience plus willingness to learn quickly.
• Depth in workload identity and access management: service identity, short-lived credentials and token exchange, and key management (SPIFFE/SPIRE, OAuth token exchange, KMS-backed signing, or equivalent).
• Strong written communication. Architecture decision records and design documents are reviewed before implementation begins on this team.
• You operate as a peer to senior engineers and leaders outside your team, not as a customer.
• Personal experience using AI tools in your own engineering work (Claude Code, Cursor, custom MCP servers, agentic side projects, etc.).