Stay Ahead of the AI-Native Revolution
Get monthly updates on new open-source repos we've added, AI-native development trends, and curated insights from the world of intelligent software engineering.
No spam, ever. Unsubscribe anytime.
What We're Reading
Spec-Driven Development with AI
Rather than relying on vague prompts to AI coding agents, the spec-driven approach establishes clear specifications as "executable artifacts" that guide implementation through four phases: Specify, Plan, Tasks, and Implement.
The specification becomes the source of truth
Welcome to Gas Town
Steve Yegge explores the new world of AI-assisted development, where code generation is cheap and abundant like gasoline, fundamentally changing how we think about software engineering.
Code is now cheap like gasoline
The Trillion Dollar AI Software Development Stack
Generative AI is revolutionizing software development, with AI coding assistants and agentic tools transforming how 30 million developers plan, code, review, and deploy software worldwide.
$3 trillion yearly GDP contribution potential
Context Graphs: AI's Trillion-Dollar Opportunity
The next trillion-dollar platform opportunity lies not in traditional systems of record, but in capturing "decision traces"—the reasoning, exceptions, and precedents behind automated decisions.
Services as Software revolution
Modular Manifolds
Exploring how constraining neural network weight matrices to mathematical submanifolds maintains tensor health during training. Introduces "manifold Muon," a variant of the Muon optimizer designed for the Stiefel manifold.
Manifold optimization for neural networks
How Netflix Built a Real-Time Distributed Graph
Part 1 of a series on building a distributed graph for real-time applications. Netflix processes over 1 million Kafka messages per second using Apache Flink, powering ads, live events, and gaming.
1M+ Kafka messages/sec at internet scale
Featured Learning Paths
Podcasts We're Listening To
Latent Space
swyx & Alessio
The Rise of AI Engineering
Deep dive into how AI is transforming software engineering workflows and the emergence of AI-native development tools.
Acquired
Ben Gilbert & David Rosenthal
Anthropic
The complete history of Anthropic, Claude, and the race to build safe, beneficial AI systems.
Practical AI
Daniel Whitenack & Chris Benson
Coding with AI Assistants
Practical strategies for integrating AI coding assistants into your development workflow.
Interesting New Tools
A2UI
Google's open protocol for agent-driven interfaces - lets AI agents safely send rich, declarative UIs
Chordio
AI-powered design reviews that catch accessibility, usability, and visual issues in Figma and browser
Doctave
Docs-as-code platform for building modern developer documentation with OpenAPI support and Git workflows
Baseten
AI inference platform to deploy and scale ML models with blazing-fast cold starts and 99.99% uptime
CC.dev
Cloud-native development environment with AI-assisted coding capabilities
Chai
BDD/TDD assertion library for Node.js and browsers that pairs with any JavaScript testing framework
Zed
High-performance code editor from creators of Atom and Tree-sitter, built in Rust with AI and multiplayer
Warp
Agentic development environment with AI agents that run terminal commands and work with MCP tools
ripgrep
Line-oriented search tool that recursively searches directories while respecting gitignore rules
Voices from the Community
Andrej Karpathy
@karpathy
"The hottest new programming language is English. AI coding assistants are fundamentally changing how we think about software development."
Simon Willison
@simonw
"Every developer needs to understand: AI coding tools are not replacing programmers. They are amplifying what good programmers can do."
Guillermo Rauch
@rauchg
"The best AI dev tools are the ones that understand your codebase context. Without context, you are just generating generic code."
Swyx
@swyx
"AI Engineering is the new frontier. We are not just prompt engineers anymore - we are building systems that think, reason, and act."
Sarah Guo
@saranormous
"The companies winning with AI are those treating it as a multiplier, not a replacement. Human judgment + AI speed = unstoppable teams."
Amjad Masad
@amasad
"We are entering the age of ambient programming. Code will be written continuously in the background, adapting to your intent in real-time."
Don't Miss an Issue
Join thousands of developers staying ahead of the AI-native development curve.
Subscribe Now