Senior AI Engineer, Agent Harness & LangChain

Relativity; Katowice; Hybrid; PLN 188,000 - 282,000; Full time

Benefits offered by this job

DTO

Parental leave

Equity program

Job summary

Relativity is a leading legal data intelligence company seeking an Advanced AI Engineer to build and extend the agent runtime. You will work with LangGraph, LangChain, and Deep Agents to create scalable, production-ready capabilities for agent orchestration and tool execution.

You will collaborate with Applied Science and aiR teams, contribute to design discussions, and mature the harness to support long-running tasks and broader protocol support ahead of high traffic.

Qualifications

• 3+ years of professional software engineering experience.
• Experience building LLM-powered or agentic systems with LangChain/LangGraph.
• Ability to write clean, testable code and contribute to design of owned components.
• Strong understanding of async Python, API/service design, and PostgreSQL.

Responsibilities

• Build and extend the agent runtime in Python using LangGraph, LangChain, and Deep Agents.
• Implement harness capabilities: streaming, tool calling, memory, session handling, and time-travel resumes.
• Develop MCP tool servers/clients, A2A interoperability, and runtime configuration registry.
• Work with multiple LLM providers and ensure production-ready delivery.
• Write clean, well-tested code and participate in design/reviews.

Skills

Python

LangChain

LangGraph

Deep Agents

Async Python

API design

PostgreSQL

CI/CD

Docker

Kubernetes

Cloud-native

Education

Bachelor's degree in CS/Engineering

Tools

Pulumi

Terraform

MLflow

OpenTelemetry

RAG

MCP

Job description

Relativity is a leading legal data intelligence company seeking an Advanced AI Engineer to build and extend the agent runtime. You will work with LangGraph, LangChain, and Deep Agents to create scalable, production-ready capabilities for agent orchestration and tool execution.

You will collaborate with Applied Science and aiR teams, contribute to design discussions, and mature the harness to support long-running tasks and broader protocol support ahead of high traffic.