engineering.
113 writings found
Latest Archives
Running Routes and Lost Code: Why LLM Transparency Matters
ChatGPT created perfect running routes using OSM data, but the actual code vanished. Why LLM systems need better transparency and code preservation.
Why I Built Laya: Open-Source System 1 for Real Production Speed
Sub-35ms open-weight decision engine with multilingual routing and honest calibration. How I rebuilt what frontier labs closed.
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.
System One Models: Why AI Automation Stayed Broken Until Now
TypeSafe's Jev model addresses the fundamental gap between chat intelligence and production automation. Here's why structured decisions matter more than raw capability.
What 1 Million Isolated Highlanders Teach Us About System Design
Why Papua New Guinea's highlands reveal fundamental truths about technological evolution, constraints, and why isolation prevents innovation.
Contributing to OpenStreetMap in 15 Minutes
Learn how to add website tags to OpenStreetMap using JOSM. A practical guide for developers who want to improve open geospatial data.
BZip3: Why Compression Algorithms Still Matter in 2024
BZip3 achieves better compression ratios than its predecessor through advanced entropy coding and Burrows-Wheeler transforms. What this means for your infrastructure.
Reading Claude's System Prompts Like Code
Anthropic publishes Claude's system prompts with version history. What we can learn by diffing them, and what's still hidden.
Anthropic's System Prompt Overhaul Shows New Legal and Safety Pressures
Fable 5.1's updated system prompt reveals how AI companies respond to legal threats, user complaints, and safety concerns through constraint engineering.
Claude Fable 5.1: When Reasoning Effort Actually Matters
Testing Anthropic's new Fable 5.1 across reasoning levels reveals a wild cost/quality tradeoff. The pelican benchmark shows reasoning might not be what we think it is.
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.
14 CPU Cores Wasted: The Real Cost of AI Scraping
How LLM training scrapers consume 20% of git.kernel.org's infrastructure, forcing difficult tradeoffs between openness and sustainability.
Why Lines of Code Still Matter When Agents Code
Coding agents are changing productivity metrics. I argue lines of code remains meaningful when measuring agent output - if the quality stays high.
Why Lines of Code Still Matter When AI Writes Your Software
AI coding agents are changing productivity metrics. We need new frameworks to measure what agents actually deliver, and what gets lost in the speed.
When AI Commoditizes Your Product: Lessons from Revenue Decline
How AI-powered alternatives are disrupting SaaS businesses and what indie developers should do about it. A case study in staying ahead of commoditization.