Senior AI Agent Engineer

Relay (Uneven Labs, Inc.)•,

6d•Remote

About The Position

Relay is hiring a Senior AI Agent Engineer on its Developer Experience team to own Relay’s internal, Slack-native AI agent. The agent works across company tools to investigate issues, answer questions, and complete operational work. This is a hands-on engineering role measured by the agent’s ability to complete tasks accurately and safely. The engineer will own and improve the agent, translating expert workflows into agent skills, decision logic, and escalation paths. They will build knowledge systems for finding and reasoning over structured and unstructured information, improve retrieval, orchestration, and verification, and measure agent performance. They will also connect agents safely to internal systems through APIs and MCP tools, choosing among deterministic code, automation, and agent behavior based on the task. The position is remote within the United States and requires dependable overlap with the teams whose workflows are being automated, including coverage of US business hours.

Requirements

At least 5 years of experience in software engineering, developer tooling, or platform engineering.
Production experience with LLM agents, including tool use, retrieval, structured outputs, and orchestration.
Experience building search, retrieval, or knowledge systems across structured and unstructured information.
Strong TypeScript or Python skills and experience building API integrations.
Experience designing evaluations or automated tests for nondeterministic software.
Familiarity with permissions, secrets, auditability, and safe handling of side effects.
Strong written communication and ability to turn an expert’s workflow into a precise specification.
Product-minded engineering approach focused on shipping AI systems into real workflows.
Care about correctness, permissions, and failure handling as much as model capability.
Ability to read traces before changing prompts, recognize when conventional software is preferable to an agent, and encode expert decision-making into testable behavior.
Comfort being the first dedicated hire for a new function.

Nice To Haves

Experience with MCP or comparable tool and context protocols.
Experience with Slack apps, support automation, or internal platforms.
Experience with entity resolution, knowledge graphs, ontology or taxonomy design, ranking, or semantic enrichment.
Experience in crypto, payments, or another high-stakes domain.

Responsibilities

Cover US business hours and maintain dependable overlap with teams whose workflows are being automated.
Own the agent backlog and prioritize work by impact, frequency, and risk.
Review real agent misses weekly and convert them into regression evaluations, fixed tools, or improved guidance.
Turn expert workflows into agent skills, decision logic, and escalation paths.
Build knowledge systems that help agents find, connect, and reason over Relay’s structured and unstructured information, with clear provenance and freshness.
Improve retrieval, orchestration, and verification, and own regressions through resolution.
Measure and report task success, correction rate, latency, and cost.
Integrate agents with internal systems through APIs and MCP tools.
Choose deterministic code, simple automation, or agent behavior according to the merits of each task.
Design permissions, side-effect controls, human approval, and failure recovery as product features.