Software Development
56 stories tagged Software Development.
Dark Code: When AI Writes Software Nobody Actually Understands
AI-generated code is shipping to production with no human comprehension. It's not a security problem—it's an organizational capability crisis.
Karpathy Skills Repo: Discipline Over Features for AI Code
Karpathy Skills Repo: Discipline Over Features for AI Code
Andrej Karpathy's lightweight GitHub repo tackles AI coding agents' behavioral problems with four principles that prioritize reliability over power.
How to Build Git Version Control Into Your Apps
How to Build Git Version Control Into Your Apps
LibGit2 lets developers embed Git functionality directly into applications. Here's what that actually looks like in practice, and why it matters.
When Your AI Agent Ships Code While You Sleep
When Your AI Agent Ships Code While You Sleep
Hearth AI founder Ashe Magalhaes builds entire products in hours using AI agents. What does this mean for engineering work and software development?
Three Hours of Debugging a File Compressor in C
Three Hours of Debugging a File Compressor in C
Dr. Jonas Birch spent 3.5 hours live-coding a file compressor in C. What the session reveals about real programming work might surprise you.
GitHub's Copilot SDK Turns Apps Into AI Planners
GitHub's Copilot SDK Turns Apps Into AI Planners
GitHub demonstrates how its Copilot SDK transforms static planning apps into dynamic AI assistants with minimal code. But what's the implementation cost?
When Software 'Works' But You Can't Trust It
When Software 'Works' But You Can't Trust It
A veteran Microsoft engineer explains the difference between software that appears to work and software that actually works—and why that gap matters.
Linear Says Issue Tracking Is Dead. Here's What's Next
Linear Says Issue Tracking Is Dead. Here's What's Next
Linear, the issue tracker beloved by engineers, just declared its own category obsolete. AI agents are changing how software gets built—for better or worse.
Anthropic's Cloud Tasks Point to 'Software Factory' Future
Anthropic's Cloud Tasks Point to 'Software Factory' Future
Anthropic's new remote task scheduling for Claude Code suggests AI development is heading toward autonomous 'software factories' running 24/7.
The Four Types of AI Agents Companies Actually Use
The Four Types of AI Agents Companies Actually Use
Most companies misunderstand AI agents. Here's the taxonomy that matters: coding harnesses, dark factories, auto research, and orchestration frameworks.
GSD Framework Tackles AI Coding's Real Problem: Choice
GSD Framework Tackles AI Coding's Real Problem: Choice
GSD, BMAD, and Superpowers take radically different approaches to AI coding. The AI LABS team breaks down when each framework actually works.
AI Is Breaking Open Source, and GitHub Won't Save Us
AI Is Breaking Open Source, and GitHub Won't Save Us
AI-generated pull requests are flooding maintainers, degrading code quality, and making open source maintenance unsustainable. Here's what's actually happening.
While Everyone Panics About AI Layoffs, Some Companies Are Hiring
While Everyone Panics About AI Layoffs, Some Companies Are Hiring
Whoop just doubled its workforce while others cut headcount. The companies hiring aggressively might understand something the cutters don't about AI's actual impact.
Fresh Text Editor Challenges Terminal Editing Status Quo
Fresh Text Editor Challenges Terminal Editing Status Quo
Fresh Text Editor brings modern IDE features to terminal environments. Policy implications for developer tools, accessibility, and open-source ecosystems.
The Engineer Who Stopped Writing Code
The Engineer Who Stopped Writing Code
Boris Cherny created Claude Code at Anthropic. Now he doesn't write any code himself. A year into AI-assisted development, what have we learned?
Effect-Oriented Programming: Making Side Effects Safe
Effect-Oriented Programming: Making Side Effects Safe
Three authors explain how effect-oriented programming brings type safety to the messy, unpredictable parts of code—without the intimidating math.
Not Every Problem Needs AI. Here's How to Tell.
Not Every Problem Needs AI. Here's How to Tell.
Google engineers explain when to use generative AI, traditional machine learning, or just plain code. The answer matters more than you'd think.
When AI Builds a Compiler in Two Weeks: What Just Changed
When AI Builds a Compiler in Two Weeks: What Just Changed
Anthropic's Claude built a 100,000-line C compiler autonomously in two weeks. IBM experts debate whether this milestone was inevitable—and what it means for developers.
AI Agents That Work While You Sleep: The Next Shift
AI Agents That Work While You Sleep: The Next Shift
Cloud-based AI coding agents now run scheduled tasks overnight. A developer built a news monitoring system in one afternoon that never sleeps.
Most Developers Using AI Are Getting Slower, Not Faster
Most Developers Using AI Are Getting Slower, Not Faster
A rigorous study found developers using AI tools took 19% longer while believing they were 24% faster. What's really happening with AI coding?