Principal Product Manager, Agent Intelligence
Core
Own the evaluation, observability, and feedback systems for AI agents to define quality, measure performance, and enforce reliability across Equinix's multi-component agent workflows.
Role type
Principal Product Manager (AI Agent Quality & Observability)
Builds
Evaluation platform, golden datasets, CI/CD quality gates, and end-to-end tracing systems for production AI agents.
Domain
AI Agents, LLM Systems, Observability, Enterprise Infrastructure
Deliverable
production ML models
Required skills
Product management for technical platforms, systems engineering background, building evaluation/observability systems for production ML/LLM, multi-component system evaluation, evaluation judgment (code-based checks, LLM judges, human review), hands-on data skills (SQL, Python, traces/logs), leading without authority
Preferred skills
LLM evaluation tools, OpenTelemetry, MCP/A2A protocols, retrieval/ranking/confidence calibration, experimentation platforms, regulated industry data governance
Technologies
Python, SQL, OpenTelemetry, MCP, A2A, CI/CD pipelines
Responsibilities
Define quality standards and measurable evals for agents, own the roadmap for offline/online evaluation and golden datasets, build trajectory-level evaluation to trace failures to components, partner on tracing/telemetry for observability, mine production traces for failure modes, lead cross-functional teams to set evaluation standards.
Seniority
Principal, hands-on IC with strategic influence
