Senior AI Engineer
Summary
Senior AI Engineer on an initial 12-month fully remote contract (Poland-based) building LLM agents for a client: typed agent frameworks (Pydantic AI, LangGraph/LangChain), MCP and multi-agent communication, data-analysis agents, plus extensibility, data security, evals, and production observability.
Our client currently seeks a Senior AI Engineer to join their dynamic team on an initial 12 month contract. This role is fully remote.
Essentials:
• Experience building agents with Pydantic AI or a comparable typed agent framework
(i.e., LangGraph, LangChain, or OpenAI Agents SDK).
• Experience with MCP and multi-agent or cross-application agent communication.
• Experience with core agentic patterns: tool calling, multi-step orchestration, context
management, guardrails, and human-in-the-loop.
• Experience designing agents that perform data analysis: exploring datasets,
generating and executing code, and interpreting results for users.
• Ability to architect as well as build: you can weigh trade-offs between frameworks,
define clean boundaries, and explain your decisions.
Required: extensibility, data, and security
• Experience designing configurable, extensible systems, for example per-client prompt
and behavior variation, plugin or skill patterns, and versioned, testable configuration.
• Understanding of data security in agent workflows: least-privilege data access, safe
handling of client data, and clear controls around permissions and failure handling.
Required: evaluation and observability
• Practical approach to evaluating LLM systems: offline evals, LLM-as-judge combined
with deterministic checks, and regression testing.
• Ability to tell whether an AI feature is working, by tracking quality, latency, cost, and
failure rates in production.
• Experience with observability for LLM systems (Phoenix, OpenTelemetry, or similar).
Skills
Sign in to read and join the discussion.