Relativity is a leading legal data intelligence company that builds technology to help users organize data, discover the truth, and act on it with confidence. As an Advanced AI Engineer on the Agent Harness team, you will build and extend our agent runtime, working hands-on with industry frameworks to ensure our AI products have a solid, scalable foundation.
Key responsibilities
- Build and extend the agent runtime in Python using LangGraph, LangChain, and Deep Agents, focusing on stateful graph agents and multi-agent orchestration.
- Implement critical harness capabilities such as streaming, tool calling, human-in-the-loop review, session memory, and checkpoint/resume functionality.
- Contribute to the extensibility and protocol layer, including Model Context Protocol tool servers and Agent-to-Agent interoperability.
- Collaborate with Applied Science and application teams to move new capabilities from experiment into production safely.
- Write clean, well-tested code and participate in design and code reviews, taking ownership of components through the full development lifecycle.
Requirements
- 3+ years of professional software engineering experience with strong, recent Python skills building production systems.
- Proven experience building LLM-powered or agentic systems with frameworks like LangChain or LangGraph.
- Solid understanding of modern async Python, API and service design, and relational data using PostgreSQL.
- Experience delivering cloud-native systems on Azure or similar platforms with CI/CD, Docker, and Kubernetes.
- Strong design instincts and the ability to contribute to the architecture of complex, scalable components.
What we offer
- The opportunity to work on distributed, cloud-native systems that process large volumes of data for high-stakes legal matters.
- A collaborative culture focused on knowledge sharing, continuous improvement, and inclusive team dynamics.
- The chance to build with cutting-edge AI technologies and scalable architectures that are actively shaping the future of legal tech.
- Professional growth through end-to-end ownership of systems across cloud and distributed environments.