Sessions in this track
Monday 16 November. 6 sessions per track, chosen and introduced by the Track Host.
10:35 Marina Sponsored Data-Driven Optimization of AI Agents: From Hand-Tuned Harnesses to Model-Harness Co-Training Jeremy Tian Founder & CPO @Lucidic AI, previously @DRW Most teams with agents in production still improve them by hand: a failure comes in, someone edits a prompt or a tool description, and the fix quietly breaks something else. 11:45 Marina Sponsored Why Optimized Application Infrastructure (not just GPUs and LLMs) Can Scale Your AI Investments Alex Berkoff Senior Solutions Engineer @Azul Systems AI adoption is mandatory, but new budget is scarce, and production AI costs are averaging 380% over pilot projections. The GPUs and models get the attention, while the systems underneath get far less. 13:35 Seacliff ABC Sponsored Building a Safe Software Factory Andrew Finch, Nima Karimipour Coding agents have made code cheap to write, and teams are shipping 10× more changes than they did two years ago. Human review was built for human speed, so the industry is moving from reviewed PRs to automated software factories. 14:45 Marina Sponsored The Context Gap: Why Correct Agents Fail Xiaowei Jiang CEO and Chief Architect @Tacnode Two agents edit different files. Git merges clean. The program is broken. No conflict was thrown, because the collision was in meaning, not in text — and business decisions are almost all semantic. 15:55 Marina Sponsored Critical Path: Nine Reliability Patterns from Processing 350 Million Payments a Day Vimal Kumar Founder and CEO @Juspay Distributing a system does not distribute away its failure modes. Replicas still depend on failover logic, shards on routing, and every safe deployment on a release gate. These few mechanisms decide whether a local failure stays local or becomes an outage, and nothing can route around them. 17:05 Marina Sponsored Next level SAP experiences with Agentic Systems Engineering Roy Van de Kerkhof Co-Founder @NovioQ SAP is a difficult system to work with, because every customer customizes SAP to fit their unique needs - and that customization is not accessible to the outside world, like agents or MCP servers.
76%
senior dev or higher
1:11
speaker ratio
60+
practitioners
QCon San Francisco 2026 is a three day conference for senior software engineers, architects and team leads. An international program committee of working engineers selects every session. Patterns and practices, not products and pitches.