QCon London 2026: AI Agents Write Your Code. What’s Left For Humans?
This matters because enterprise architecture decisions around AI, data, and platform engineering define long-term competitiveness and operational efficiency.
QCon London 2026: AI Agents Write Your Code. What’s Left For Humans?
Hannah Foxwell began her QCon London 2026 talk by noting that the long-sought velocity in development has arrived, but the industry is unsure how to use it. She set aside the technical details of agentic coding, focus...
Editorial Analysis
The real question isn't whether AI agents will write code—they already do—but whether we're structuring our data platforms to support them. I've seen teams treat agentic coding as a pure development acceleration tool, then scramble when they realize data quality, lineage tracking, and observability become critical bottlenecks. These agents need clean schemas, consistent metadata, and reliable data contracts to function reliably. This pushes us toward stronger data governance and modular architecture. The architectural implication is clear: teams investing now in dbt frameworks, data catalogs, and automated testing infrastructure will extract real value from AI coding agents. Those building monolithic pipelines or skipping lineage documentation will find agents generating more technical debt than velocity. The broader trend is consolidation—AI tools force us to finally address infrastructure debt we've deferred for years. My recommendation: audit your metadata layer and data quality practices before adopting agentic development. The bottleneck won't be code generation; it'll be data reliability.