Yusuf Olokoba, founder of Muna, details a compiler-based approach to transform Python AI functions into self-contained native binaries. This talk explores the technical pipeline, including custom AST-based tracing, type propagation, and the strategic use of LLMs for code generation, enabling a universal, OpenAI-style client for running any model on any platform.
Siddharth Bidasaria from Anthropic shares the origin story of Claude Code, from a simple internal terminal app to a powerful coding agent. He discusses the team's core philosophy of 'letting the model cook,' the evolution of agentic capabilities, the critical role of verification and testing, and the future of complex, multi-agent systems.
Amjad Masad, founder of Replit, joins a16z to discuss the rise of AI agents that can now plan, reason, and code for hours. He explains how reinforcement learning and verification loops unlocked long-horizon reasoning, why AI is advancing fastest in verifiable domains like code, and debates whether "good enough" AI might be a local maximum that blocks the path to AGI.
OpenAI's Greg Brockman and Thibault Sottiaux explore the evolution of AI-powered software development, from the first coding sparks in GPT-3 to today's GPT-5 Codex. They detail the concept of the "harness," the rise of agentic coding, breakthroughs in automated code review, and how AI agents capable of running for hours on complex refactoring will reshape the future of engineering.
Jay Alammar, Director at Cohere, discusses the practical adoption of Large Language Models in the enterprise. He covers the evolution of Retrieval-Augmented Generation (RAG) from a simple anti-hallucination tool to complex, agentic systems, the critical role of evaluation as intellectual property, and future trends like text diffusion and the increasing capability of smaller models for specialized business tasks.
Birgitta Böckeler from Thoughtworks provides a comprehensive overview of the current AI tooling landscape for software delivery. The talk focuses on the architecture and capabilities of modern coding assistants, differentiating between the underlying model and the crucial role of tooling in providing context. It explores what works today, the promising but challenging frontiers like testing and refactoring, and offers a practical guide to the major tools in the ecosystem.