The wiki

Dust cofounder and engineer working on workplace AI, agent collaboration and formal verification, with earlier research contributions to neural theorem proving.

Part 3 of the ChatGPT voice preview tests the gap between conversational control and precise audio behavior: recognizing an illusion, producing a tuning note, copying pitch, and thinking before speaking.

AI researcher whose work spans children's AI literacy, creative programming with families, and self-improving research agents.

Neo4j developer relations leader whose AI teaching focuses on graph retrieval, agent memory, and helping developers understand where technology applies.

Founder of tldraw and creator of Perfect Freehand, Ruiz develops canvases for visual work with AI, from annotated prototypes to local agent scripting.

Software engineer and writer whose work spans compiler-derived code intelligence, AI coding assistants, and systems for coordinating coding agents.

Coding agents move the work toward asking questions, supervising execution, and integrating changes. The promise comes with a learning curve—and a bottleneck that more code cannot solve.

OpenAI’s reasoning project team associated with o1 and the shift toward spending more computation on inference.

Stripe’s financial infrastructure connects payment prediction, AI usage billing, agent checkout, and developer services.

Part 1 follows the path from production outages to dependable model interfaces: constrained decoding, tool execution, refusals, schema design, evaluation, and model selection.

Playground founder and image-generation research coauthor whose work combines model development with visual editing tools.

Systems engineer whose work on Fullstack-Bench and Convex’s MCP server examines how feedback, code, and abstractions shape autonomous coding.

Code execution, open weights and long context expanded what developers could build. Evaluation, deployment control and reliable behavior determined whether those capabilities could become useful products.

Cloudflare engineer and PartyKit founder working on durable AI agents, sandboxed code execution, and software that lets people and agents act on the same work.

An AI music company whose products evolved from generative audio and prompted songs into personalized models, a music workstation, and models developed with music-industry partners.

The email business evolved from on-demand AI writing tools into contextual reply drafting, then became part of a broader productivity suite under the Superhuman name.

Developer of ChatWithPDF and Archon, whose work spans document retrieval, computer-use model training and the economics of AI applications.

Shawn Wang, known as swyx, writes about AI engineering, co-founded the AI Engineer conference series, and experiments with software-development agents.

EntreConnect co-founder and AI community organizer, identified by AWS Events in 2026 as MiniMax’s Head of North America.

How generated textbooks, rewritten web pages and synthetic labels support capable models that cost less to run and can stay on the device.

Toronto-based developer of model-specific AI inference chips, whose HC1 hard-wired Llama 3.1 8B and whose technology became the subject of an AMD acquisition agreement in August 2026.

Computer vision researcher whose collaborative work addresses control, motion coherence and customization in image and video generation.

Medical AI researcher and Sophont co-founder working on open multimodal foundation models, with earlier contributions to MedARC, neuroAI research, and open-source machine learning.

AI can generate a starting interface. Choosing a direction, making its behavior coherent, and developing a distinctive product still demand design judgment.

Cosine’s software agent learns from reconstructed development work: finding relevant code, planning changes and recovering from mistakes. Its experience shows why better patches depend on more than code generation.

Part 6 of 6 of Latent Space’s ICML 2024 recording: faster robot policies, Chelsea Finn’s approaches to learning from language and experience, and Younghyo Park’s case for automatic environment shaping.

An AI engineering and transformation firm built around output-based billing, engineer compensation, and shared infrastructure for coding agents.

AI growth and developer education practitioner whose software factory argument extends autonomy beyond coding to the whole development lifecycle.

Engineer on Anthropic’s Claude Code team whose public work explains how to adapt agent context, verify generated software, and keep human judgment involved.

How requirements, mutable harnesses and agent security fit together in a conversation about the next era of coding agents.