AI-Native Development of Complex Enterprise Software

AI-assisted software development is moving past smart auto-complete and quick prototypes toward a structured, human-mediated, multi-agent workflow that spans the full codebase and the entire end-to-end development process. This workshop-style session with Kerry Champion examines how spec-driven development (SDD) and human-in-the-loop multi-agent workflows address the hard problems of applying these techniques to enterprise-class projects: context engineering, agent memory management, code stability, and the division of labor between agents and humans. The evening is split into two talks. Part 1 takes a top-down view using the open source BMAD framework, covering multi-agent orchestration, context sharding and bounded-context design, structured artifact pipelines, and applying these methods to mixed greenfield and brownfield codebases. Part 2 takes a bottom-up view using CodeSpeak, the LLM-powered programming system from the creator of Kotlin, which defines a higher-level specification layer above traditional programming languages so edits to the spec drive reliable, traceable updates to the underlying code. The session is aimed at engineers, technical leaders, AI tooling practitioners, and founders interested in agentic systems, developer tooling, and large-scale AI-assisted development. Doors open at 18:00, talks run from 18:30 to 20:00, and networking follows until close at 21:00.