Senior Platform Engineer 2, AI Tooling

Responsibilities

• Architect the internal AI platform: MCP servers, agent customizations, agentic workflows, engineering harness, and integrations that make AI the default way engineers get work done at Drata
• Design and ship custom agent skills, subagents, hooks, and plugins tuned to specific engineering workflows (code review, test generation, PR triage, on-call, release notes, migrations)
• Build end-to-end agentic workflows that automate code generation, verification, testing, and delivery
• Design async, recoverable, long-running agent workflows using Temporal
• Lead the rollout and customization of AI coding tools across the org
• Wire AI into CI/CD: automated PR review, test generation, doc generation, migration agents, bug triage agents, autonomous coding agents in pipelines
• Build the governance layer: usage telemetry, cost tracking, audit trails, security policy enforcement
• Design and run evals to measure which prompts, skills, and agents save time vs. cost time
• Measure adoption and impact with engineering productivity frameworks
• Partner with Security, Legal, and Compliance to keep AI usage inside Drata's customer-trust boundaries
• Mentor engineers, set technical direction, and review designs across the org
• Stay current with the AI coding tool landscape
• Participate in design and coding activities, write reusable and testable code, deliver in an agile team environment
• Participate in design reviews, post-mortems, and on-call for the AI Tooling platform

Requirements

• 7+ years of experience as a software engineer building production systems and platforms
• 1+ years of hands-on experience customizing or extending AI coding tools in a real engineering context: agent skills, subagents, hooks, plugins; agentic IDE rules and extensions; MCP servers; or similar
• Daily user of AI coding tools
• Deep working knowledge of MCP (Model Context Protocol): authoring servers, designing tool interfaces, scoping permissions, handling auth
• Strong working knowledge of the Anthropic API (preferred) or equivalents: tool use, structured outputs, prompt caching, batch API, streaming
• Solid prompt engineering experience for agentic workflows — multi-step engineering tasks with tools, verification, and recovery
• Track record of building agentic workflows that span the full coding lifecycle: planning, generation, verification, testing, and delivery
• Experience integrating AI into CI/CD pipelines: GitHub Actions, automated PR review bots, agentic test generation, background coding agents
• Strong backend and systems background: TypeScript, NodeJS, or Python in production, REST and event-driven architectures, CI/CD systems
• Comfort building developer-facing tooling: CLIs, GitHub Apps, internal portals
• Working understanding of the security and governance side of AI dev tools: code and data egress policies, secrets in prompts, prompt injection risk inside repos, audit trails for AI-driven changes
• Understanding of LLM cost and latency tradeoffs as they show up in dev workflows: token caching, batch jobs, async agents, model selection
• Strong skills in the Drata core stack: NodeJS, TypeScript, Temporal, MySQL, Git, REST
• Experience building in AWS
• Familiarity with engineering productivity measurement (DORA, DX, SPACE) and how to instrument it
• Track record of owning ambiguous platform problems end-to-end and shipping them
• Comfortable making architecture calls under uncertainty in a space that changes monthly
• Experience mentoring engineers and unblocking other teams
• Outstanding ability to negotiate difficult tradeoffs (quality vs. speed, build vs. buy, mandate vs. paved road)
• Excellent written communication: design docs, RFCs, post-mortems

Nice to Have

• Published agent skills, plugins, or open-source AI dev tool contributions
• GitHub App development at scale
• Terraform