Thomson Reuters
Lead Software Engineer, AI
$127K–$237K
United States of America, Eagan, Minnesota +3 · Hybrid · Full-time · Posted Oct 14, 2026
- Location
- United States of America, Eagan, Minnesota
United States of America, Frisco, Texas
Canada, Toronto, Ontario
United States of America, Ann Arbor, Michigan - Experience
- No number of years stated; extensive experience is requested.
About the job
Leads architecture and hands-on development of scalable AI agent infrastructure, tooling, and production reliability practices for legal AI systems.
What you'll do
- Set technical direction across agent infrastructure, internal tooling, and production reliability initiatives.
- Build production AI agent systems and observability and evaluation practices, and improve deployment safety and release speed.
- Lead design decisions, mentor senior engineers, and align architecture with product and engineering priorities.
What they ask for
- Extensive experience architecting and shipping production software, including hands-on AI/LLM product or agentic-system development.
- Direct experience with modern LLM APIs, agent frameworks/SDKs, or agent-to-tool integration protocols.
- Strong backend fundamentals in distributed systems, service architecture, and cloud infrastructure; experience with production reliability and observability.
- Track record setting direction across concurrent initiatives and mentoring senior engineers.
Nice to have
- Experience leading architecture for developer-facing platforms, plugin/extension systems, or integration ecosystems; familiarity with progressive delivery at scale.
- Experience with managed agent runtime infrastructure; legal, regulated, professional-services, or fiduciary-domain expertise is a plus.
Pay
$127,400 USD - $236,600 USD (base salary); $140,000 CAD - $175,000 CAD (base salary)
Tools
modern LLM APIs, agent SDKs, AWS Bedrock AgentCore
AI focus
Architect and build production AI agents and agent infrastructure; develop tracing, automated evaluations, quality/drift detection, and safer production deployment practices. The team also uses AI-native agentic coding tools.
Model-written summary · the original posting is authoritative