developer-tools.
33 writings found
Latest Archives
ChatGPT Work Generated Running Routes, Then Lost The Code
GPT-6 created GPS routes from my address using OSM data, but the actual code disappeared after thread compaction. What this reveals about LLM transparency.
ChatGPT Work: When AI Gets Real Internet Access
OpenAI's new Work mode gives Claude a powerful upgrade: real internet access, browser automation, and persistent file systems. Here's what it means.
Why Tokenmaxxing Misses the Mark for AI Coding
Token optimization in AI agents creates perverse incentives. Real value comes from measurable outcomes: PR merges, release velocity, and actual developer productivity gains.
GitHub Copilot Slash Commands Transform Developer Workflows
Explore how slash commands in GitHub Copilot app enable faster planning, implementation, and collaboration without breaking developer focus.
LLM 0.32: When a CLI Tool Becomes an Agent Platform
Major LLM release adds reasoning traces, server-side tools, and content-addressable logging. What this means for building AI systems.
Stateless MCP has recaptured my interest
MCP 2.0 simplifies agent tooling with stateless design. I built three projects this week exploring why this matters for secure, scalable LLM applications.
Why Your Laptop Just Became Production for AI Agents
AI agents are reshaping the SDLC. Runtime isolation, governance, and the three-layer security model are now table stakes for shipping safely.
How GitHub Cut Copilot Review Cost 20% by Fixing Agent Instructions
GitHub's Copilot code review team found that swapping tools wasn't enough. Rewriting agent instructions around reviewer workflows cut costs by 20%.
When AI Has No Source: Implications for Developers
Exploring what it means for developers and the AI industry when models generate content with no traceable source material.
How GitHub Cut Copilot Review Cost 20% by Rewriting Tool Instructions
GitHub's Copilot code review team found that swapping tools wasn't enough. Rewriting agent instructions around pull request workflows cut costs by 20%.
When AI Has No Source: Implications of Empty Context
Exploring what it means for developers and the AI industry when models operate with no grounding context or source material at all.
When AI Has No Source: Implications of Empty Context
What happens when an AI blog post has no source material? A look at what empty context means for developers building on top of LLMs.
Why LSP Servers Finally Make Copilot CLI Actually Smart
Copilot CLI reverse-engineers code badly. Here is how LSP servers fix that.
GitHub Copilot CLI: Why Two Modes Matter More Than You Think
Interactive vs non-interactive in Copilot CLI isn't just about speed. It's about fundamentally different ways of thinking through problems at the terminal.
Why I Actually Care About Markdown Now (And You Should Too)
Markdown isn't just formatting syntax. It's the quiet infrastructure of developer communication that makes or breaks your open source presence.