Engineering AI Systems

Most teams have shipped an AI demo. Far fewer have shipped AI systems that work reliably in production — at scale, under real constraints, and with real consequences.

Engineering AI Systems at QCon San Francisco 2026 is for senior engineers and technical leaders building applied AI beyond the prototype stage. The track focuses on the engineering challenges that emerge when AI becomes part of critical systems: reliability, evaluation, infrastructure, autonomous workflows, and operating non-deterministic software in production.

Every session comes from practitioners running real AI systems at scale.

Topics include:

• AI-Native Engineering— How agentic coding systems are reshaping software development workflows for hundreds of thousands of engineers.
• Reliability & AI Evaluation— Practical approaches to measuring, monitoring, and improving AI reliability when clean ground truth does not exist.
• Applied AI in Regulated Domains— Lessons from deploying AI in finance, healthcare, legal, and other environments where mistakes carry operational and legal risk.
• Enterprise Adoption & Autonomous Workflows— How organizations integrate AI agents into business-critical workflows companies depend on every day.
• Inference at Scale & AI Infrastructure— The infrastructure, serving, and operational decisions required to run AI systems handling hundreds of millions of requests.

This track is designed for teams already past the “should we use AI?” conversation and focused on “how do we make AI systems actually work?” Attendees will leave with concrete lessons, production-tested patterns, and a clearer understanding of the emerging AI engineering stack.