Principal Enterprise Data Engineer
Core
Design, build, and deploy the 'harness' environment that steers AI coding agents to produce correct, maintainable, and well-architected software output, shifting the bottleneck from code generation to verification and trust.
Role type
Principal Enterprise Data Engineer (Harness Engineering)
Builds
Production-grade engineering environments, agent instruction files, reusable skills, architectural rules, feedback sensors (linters, static analysis), and quality gating mechanisms for AI-generated code.
Domain
AI Engineering / Software Development Lifecycle / Autonomous Agents
Deliverable
production ML models | product features
Required skills
Software architecture and design, AI coding agents (Claude Code, Codex), engineering tooling (linters, static analysis, CI/CD), spec-driven development, agent orchestration, consumer/contract testing (Pact), security guardrails for autonomous agents, observability engineering.
Preferred skills
Experience with AGENTS.md conventions, fitness functions, deterministic vs inferential controls, quality-gating processes, mentorship.
Technologies
LLMs, CI/CD pipelines, containerized environments, Pact, static analysis tools, linters.
Responsibilities
Design and evolve feedforward guides and architectural rules for agents; build feedback sensors and verification loops; define quality gating and release criteria for agent-produced work; establish LLM testing infrastructure; run the steering loop to prevent agent mistakes; improve observability of agent work; partner with product teams to enforce specifications; mentor engineers on harness practices.
Seniority
Principal, hands-on IC with strategy & mentorship