Hybrid opportunity at
ElasticPrincipal AI Engineer - Context - Agents and Context
This hybrid, full-time principal AI engineering role is based in the United Kingdom and sits on the Context Engine team. That team builds the knowledge layer…
Career Tools
About This Role
This hybrid, full-time principal AI engineering role is based in the United Kingdom and sits on the Context Engine team. That team builds the knowledge layer AI agents use when they work with enterprise data held in Elasticsearch. Knowledge is pulled from many origins and organized into a structured AI index, exposed to agents via public APIs, MCP tooling and framework connectors, and refined using traces from real agent activity.…
Job Description
This hybrid, full-time principal AI engineering role is based in the United Kingdom and sits on the Context Engine team. That team builds the knowledge layer AI agents use when they work with enterprise data held in Elasticsearch. Knowledge is pulled from many origins and organized into a structured AI index, exposed to agents via public APIs, MCP tooling and framework connectors, and refined using traces from real agent activity. Agents that use the layer include Elastic's Agent Builder, Claude Code and third-party frameworks such as LangChain.
The engineer is responsible for the full feedback cycle that improves the product once it is in use. That means studying the behavior of extraction jobs, retrieval tools and memory components in production, using offline evaluations, customer feedback and telemetry to find failure modes, then fixing and proving those fixes. The role also sets the method for changing agents and skills safely, including versioning, staged and shadow evaluation, regression coverage and guardrails that protect customers.
Telemetry design is a major part of the job. Decisions cover which signals to log from traces, tool calls and retrieval runs, where those records land in Elasticsearch, and how they drive dashboards and evaluation work. Working with data scientists, the engineer agrees on golden datasets and evaluators that gate quality, latency and cost. Code reviews, design reviews, mentoring and technical proposals are also expected.
The codebase is TypeScript and is developed in the open. The posting describes a distributed, asynchronous team that includes data scientists, backend engineers, product managers and UX designers. The employer describes itself as a search AI company used by more than half of the Fortune 500.
Responsibilities
- Take charge of the live improvement cycle for Context Engine, from spotting failures to proving fixes
- Define safe iteration practices for agents and skills, including versioning, rollout and regression coverage
- Design telemetry from agent traces, tool invocations and retrieval, and connect it to evaluation and dashboards
- Set evaluation strategy with data scientists, including golden datasets and quality, latency and cost gates
- Review designs and pull requests, mentor engineers in evaluation-driven development, and write technical proposals
- Run product experiments end to end, from instrumentation to interpreting results
Requirements
- A decade or more in software engineering, recently spent shipping and running AI-driven products
- Track record of evaluation-driven improvement, diagnosing agent behavior from traces and user feedback
- Practical experience creating stateful agents with memory and safely updating prompts, skills and tool logic in production
- Know-how with MCP, including publishing public servers and tools
- Background instrumenting AI systems and turning that data into engineering choices
- History of evolving data models and public interfaces while keeping external users unaffected
- Solid server-side skills in Python or TypeScript across service APIs, stateful flows, pipelines and production systems
- Able to collaborate as an equal with scientists, engineers and PMs, and to articulate trade-offs
Qualifications
- Exposure to frameworks like LangGraph, CrewAI or the Claude Agent SDK
- Experience with agentic retrieval and knowledge representation
- Practical Elasticsearch experience
Core Skills
Benefits
- Pay set by the scope of the work rather than past salary
- Medical coverage for staff and their families in many regions
- Choice of location and working hours for many positions
- Generous annual vacation allowance
- Matching for charitable giving and service of up to $2,000 in local currency equivalent, plus up to 40 hours a year for volunteer projects
- Parental leave of at least 16 weeks
Frequently Asked Questions
Answers are based only on the employer’s listing; where it doesn’t say, neither do we.
Is this role hybrid?
Yes. The posting lists the work arrangement as hybrid and the location as United Kingdom.
What is the salary range?
The posting gives a typical starting base salary range of 104,800 to 165,700 EUR. It states there is no variable compensation component.
What programming skills are needed?
Strong backend engineering in Python or TypeScript is required. The codebase is TypeScript and is developed in the open.
What experience is required?
Ten or more years of software engineering experience, including recent work shipping AI-driven products on production traffic.
Sample Interview Questions
AI-generated questions tailored to this specific role — a preview of the full practice set.