Elastic

Principal AI Engineer - Context - Agents and Context

Elastic
Apply
1 month ago
Poland (city not specified)Staff+

Responsibilities

  • Own the production improvement loop for extraction automations, retrieval tools, memory, agents, and skills.
  • Define safe versioning, rollout, regression coverage, staged evaluation, shadow evaluation, and guardrails for agent behavior.
  • Design telemetry for agent traces, tool calls, knowledge retrieval, evaluation, dashboards, and feedback loops in Elasticsearch.
  • Partner with data scientists on golden datasets and evaluators for quality, latency, and cost.
  • Run product experiments end to end, including instrumentation, unattended execution, and result interpretation.
  • Review designs and pull requests, mentor engineers in eval-driven development, and write technical proposals that shape the roadmap.

Requirements

  • 10+ years of software engineering experience, including recent experience shipping and operating AI-driven products on real production traffic.
  • Track record of diagnosing agent or LLM behavior from traces and user feedback, designing evaluations, shipping fixes, and measuring outcomes.
  • Direct experience building agents with state and memory and safely iterating on prompts, skills, and tool behavior in production.
  • Familiarity with MCP, including exposing public MCP servers and tools.
  • Experience designing telemetry for AI systems and using it to guide engineering and product decisions.
  • Experience building products with public APIs and evolving data models without breaking external contracts.
  • Strong backend engineering skills in Python or TypeScript, including APIs, stateful workflows, data pipelines, and production services.
  • Ability to collaborate with data scientists, engineers, and product managers and communicate technical trade-offs clearly.
  • Bonus: experience with LangGraph, CrewAI, Claude Agent SDK, agentic retrieval, knowledge representation, or Elasticsearch.

Benefits

  • Base-salary compensation with no variable compensation component.
  • Health coverage for employees and families in many locations.
  • Flexible locations and schedules in a distributed company.
  • Generous vacation days.
  • Matching donations up to $2,000 or local-currency equivalent.
  • Up to 40 hours annually for volunteer projects.
  • At least 16 weeks of parental leave.

Tech Stack

ElasticsearchPythonTypeScript
Elastic

About Elastic

10,000+ employees

Elastic builds the Elasticsearch-based platform for enterprise search, observability (logging, metrics, APM), and security (SIEM/endpoint). It sells subscriptions and Elastic Cloud managed services with usage-based pricing, combining open source and proprietary features through Elasticsearch and Kibana. Founded in 2012 and headquartered in San Francisco, Elastic is publicly traded on the NYSE and is used by more than half of the Fortune 500.

Contact me