Senior AI Engineer - Apm Experiences

Datadog Datadog · Enterprise · Boston, MA +2 · Remote · Dev Eng

Senior AI Engineer role focused on building and shipping LLM and Agent-based features for Datadog's APM product. The role involves end-to-end development, from problem discovery and prototyping to production deployment and scaling, with a strong emphasis on creating autonomous agents that can debug, optimize, and monitor application performance.

What you'd actually do

  1. Shape AI experiences for APM. Design and ship LLM/agentic workflows that analyze traces, metrics, logs, and other telemetry to generate diagnoses, explanations, and guided fixes.
  2. Own the full loop. Prototype quickly, define success metrics and evals, run experiments, iterate, and ultimately productionize for scale and reliability.
  3. Build robust agent systems. Develop tools, retrieval and planning strategies, and guardrails; manage prompts/evals; design fallbacks and human‑in‑the‑loop paths.
  4. Integrate with Datadog’s platform. Leverage surfaces like Trace Explorer, Service Catalog, monitors, and workflows to deliver end‑to‑end value in the APM UI.
  5. Raise the bar on engineering. Write performant, maintainable backend code, own services in production, and improve reliability for high‑throughput, low‑latency data systems.

Skills

Required

  • 4+ years building backend or real-time ML systems
  • Proven experience delivering LLM/agent features to production
  • Comfortable owning user journeys, iterating from prototype → alpha → GA, and measuring impact with clear product metrics
  • Demonstrated ability to use AI coding tools in day-to-day workflows and validate, critique, and refine AI-generated output
  • Solid grasp of the ML lifecycle (task definition, dataset collection, modeling, evaluation, deployment, iteration) and statistics (experiment design, confidence intervals)
  • Experience choosing/modeling the right technique for the job
  • Fluency with offline/online evals for AI systems
  • Experience with microservices performance: tracing, latency breakdowns, concurrency, and resiliency patterns
  • Proficient in Go, Java, or Python
  • Strong API/service design
  • Production ops (monitoring, alerting, on‑call rotation)

Nice to have

  • Hands‑on with distributed tracing stacks (OpenTelemetry/Datadog APM), profilers, and logs/metrics pipelines
  • Exposure to planning/agent frameworks, tool‑use orchestration, RAG, and retrieval/indexing for observability data
  • Familiarity with SLO/SLA practices and incident response

What the JD emphasized

  • Proven experience delivering LLM/agent features to production
  • Comfortable owning user journeys, iterating from prototype → alpha → GA, and measuring impact with clear product metrics
  • Fluency with offline/online evals for AI systems; can build reliable golden sets and automatic regressions

Other signals

  • LLM-based features
  • Agent-based features
  • Production deployment
  • End-to-end ownership