Lead Software Engineer, AI
HybridToronto, Ontario, Canada or Ann Arbor, Michigan, United States
Job Summary
Architect production AI agent systems on modern LLM APIs and orchestration infrastructure, scaling infrastructure, tooling, and observability for reliable multi-agent operations. Set technical direction across core agent infrastructure, internal tooling, and production reliability, driving cross-cutting design reviews and RFCs to maintain system coherence. Design agent observability and evaluation practices including tracing, automated evals, and drift detection to catch regressions before customer impact. Mentor senior engineers on AI-native, agentic development practices while partnering with leadership to translate strategic priorities into concrete sequencing decisions.
Required Qualifications
- Extensive experience architecting and shipping production software systems
- hands-on experience building AI/LLM-powered products or agentic systems
- Direct experience building with modern LLM APIs, agent frameworks/SDKs, or agent-to-tool integration protocols
- A track record of setting technical direction across multiple concurrent initiatives
- Strong backend engineering fundamentals, including distributed systems design, service architecture, and cloud infrastructure
- Experience designing for production reliability and observability: monitoring, evaluation pipelines, incident response, and quality regression detection
- Demonstrated ability to mentor senior engineers and influence technical culture and practices
- Comfort operating with high autonomy on a fast-moving, senior-heavy team that builds with AI-native/agentic engineering practices
Desired Qualifications
- Experience leading architecture for developer-facing platforms, plugin/extension systems, or integration ecosystems
- Familiarity with progressive delivery practices (feature flagging, phased rollout, blue/green deployment) at scale
- Experience pushing a team toward elite software delivery practices: progressive delivery, shift-left quality/testing, and the kind of engineering discipline reflected in metrics like deployment frequency, lead time, and change failure rate
- Experience with managed agent runtime infrastructure (e.g., AWS Bedrock AgentCore or similar)
- Domain expertise in legal, regulated, professional-services, or other fiduciary domains
Hiring someone like this?
Get your role in front of qualified candidates on Sorce.