Manager I, Engineering - Codegen

Datadog Datadog · Enterprise · Boston, MA +2 · Remote · Leadership

Manager I, Engineering for Datadog's CodeGen team, focusing on building production systems that use AI to understand, generate, and safely modify code. The role involves leading a team, driving technical direction, partnering with cross-functional teams, and ensuring production readiness for AI-enabled code generation features.

What you'd actually do

  1. Lead a team of engineers responsible for building production systems that enable code understanding and automated code changes — from parsers and telemetry ingestion to model serving, evaluation, and PR automation.
  2. Drive technical direction and execution: set a clear roadmap, prioritize work, remove blockers, and raise quality and reliability bar for codegen features (safety, correctness, performance).
  3. Partner with product, applied research, infra/SRE, security, and other engineering teams to ship end-to-end experiences that safely apply model outputs (PRs, CI checks, automated apply flows).
  4. Build and maintain robust evaluation and monitoring pipelines (offline and online) to measure model quality, drift, and downstream correctness of code changes.
  5. Own hiring, performance development, 1:1s, and career growth for your reports; grow a high-performing, inclusive team.

Skills

Required

  • software engineering
  • applied science
  • engineering LLM-based systems in production
  • managing small teams of software engineers and/or applied scientists
  • delivering high-quality products
  • Python
  • Go
  • machine learning theory
  • statistics
  • fundamentals
  • communication abilities
  • collaborative mindset
  • working in cross-functional teams
  • proactive approach
  • continuous learning
  • innovation
  • validate, critique, and refine AI-generated output

Nice to have

  • AI coding tools in day-to-day workflows

What the JD emphasized

  • building production systems that enable code understanding and automated code changes
  • ship end-to-end experiences that safely apply model outputs
  • build and maintain robust evaluation and monitoring pipelines
  • engineering LLM-based systems in production
  • safely change real code
  • agentic remediation
  • automated fixes
  • automated apply flows
  • safely apply model outputs

Other signals

  • building production systems that enable code understanding and automated code changes
  • ship end-to-end experiences that safely apply model outputs
  • build and maintain robust evaluation and monitoring pipelines
  • engineering LLM-based systems in production