Engineering

Agent Harness Engineer

Hybrid, Salt Lake City, Utah · Full-time · Reports to top executives

The harness is the part of Liaison that decides what an agent does, how it calls tools, which model handles which task, and whether it did the job well. We own it, and it is the core of the product. You will build with our senior harness engineer and take real pieces of it as your own.

Apply for this role

About Liaison

Liaison is an AI infrastructure and data company. Our platform connects to the systems a business already runs on, builds a unified view of its operations, and deploys agents that act on that context the way the owner would, bringing forward only the decisions that require a person. Our work spans agent engineering, model training outputs for small language models (SLMs), and frontier-level research, and we build our agent runtime and integrations in-house.

What month 12 looks like

You own a meaningful part of the harness, such as a set of evals, a family of tools or a routing rule set, and it has measurably improved agent reliability.

What you will do

Build and maintain tool integrations that agents call
Write evals and test cases from real agent runs, including the failures
Help tune model routing using cost and quality results
Work on computer use for tools that have no API
Debug agent failures from logs and fix the cause
Review your own and others' changes

Required

Has shipped code that calls LLM APIs in a production application
Has written automated tests or evals for non-deterministic output
Comfortable debugging across services and logs
Has worked on a team with code review

Preferred

Has built tool calling or browser automation for an agent
Has worked on a data pipeline or backend service at an early-stage company

Interview process

A 20-minute screening call, a paid work sample, then a final conversation with our CEO.

Apply

Tell us who you are and send your résumé or your LinkedIn profile. Everyone who applies hears back.