Principal Engineer, Agentic Engineering

Pinterest Pinterest · Consumer · Chicago, IL · Infrastructure and SRE

Principal Engineer role focused on building the foundational platform and capabilities for AI agents to operate within Pinterest's software development lifecycle. This includes architecting agent runtimes, memory systems, orchestration frameworks, evaluation pipelines, and governance models to enable 3,000+ engineers to build alongside AI agents safely and at scale. The role involves shipping production systems, owning developer velocity measurement, and serving as a technical advisor to leadership.

What you'd actually do

  1. Define and drive Pinterest's technical strategy for agentic engineering — building the platform where AI agents write code, review pull requests, debug production incidents, and orchestrate development workflows across 3,000+ engineers
  2. Architect and ship production systems — modular agent runtimes, persistent memory, multi-agent orchestration (planner, executor, reviewer, tool agents), sandbox isolation, evaluation pipelines, governance (trust boundaries, code generation policies, credential protection, audit trails), and enterprise-grade security
  3. Reimagine Pinterest's SDLC — transforming coding, testing, reviewing, deploying, and monitoring into an AI-native lifecycle where humans move from "in the loop" to "on the loop"
  4. Own developer velocity measurement — telemetry, benchmarking, and evaluation systems that track agent quality, reliability, cost, and engineering throughput, building structured feedback loops that compound toward order-of-magnitude velocity gains
  5. Serve as technical force multiplier across Platform, Infrastructure, and Engineering Acceleration — integrating AI-native workflows with Pinterest's monorepo tooling, CI/CD pipelines, observability, and code quality systems while providing architectural direction, making disciplined build-vs-buy decisions, and setting quality standards through reference implementations

Skills

Required

  • 12+ years of software engineering experience
  • deep expertise in building developer platforms, internal tooling, or engineering infrastructure that measurably changed how engineers work at scale
  • demonstrated hands-on experience designing and shipping agentic AI systems in production
  • knowledge and context engineering
  • persistent memory architectures
  • multi-agent orchestration

Nice to have

  • experience with Anthropic, OpenAI, AWS, Google
  • contribute to open-source agentic frameworks

What the JD emphasized

  • architect and ship production systems
  • evaluation pipelines
  • governance
  • enterprise-grade security
  • AI-native lifecycle
  • developer velocity measurement
  • agent quality, reliability, cost
  • integrating AI-native workflows
  • architectural direction
  • build-vs-buy decisions
  • quality standards
  • hands-on experience designing and shipping agentic AI systems in production
  • persistent memory architectures
  • multi-agent orchestration

Other signals

  • AI agents writing code
  • AI agents reviewing pull requests
  • AI agents debugging production incidents
  • AI agents orchestrating features
  • foundational platform for AI agents
  • agent runtimes
  • memory systems
  • orchestration frameworks
  • evaluation pipelines
  • governance models
  • developer velocity measurement
  • integrating AI-native workflows