Developers saving 10+ hours a week with AI fell from 68% to 26%. The bottleneck moved.
Atlassian published its State of Developer Experience 2026 this week, and one number in it stopped...
Tutorials
I taught my local model taste. Here's the prompt that banned the purple gradient.
Every AI-built landing page looks the same: indigo gradient, Inter, three cards with icons in circles. Onyx 3 is my attempt to fix that in a model that runs on your own machine, pl...
One Markdown File Can Ruin Your Coding Agent. I Checked What Mine Actually Reads.
A while back I hid one sentence inside a log file and handed the file to my coding agent. It ran the...
I Read Codex's Source to Find When It Auto-Compacts. It's 90%, and You Can't Turn It Off
Originally published at its-tahir.com. Codex users keep asking when auto compaction fires and why...
Caret: an open-source Cursor class AI code editor
I wanted the Cursor experience without shipping a second copy of Chromium and without a monthly fee...
# The 0 / 5 / 20 incident learning curve: watching an agent get better without retraining
"It learns over time" is the most common claim made about AI agents with memory, and the hardest one...
Does Having Nothing Make You Better at Shipping AI Products Than a Fortune 500?
This question came up in Chinese AI developer circles recently, sparked by a talk Wang Jian gave at...
Why Website Migrations Can Hurt SEO Even When URLs Stay the Same
Keeping the same URLs during a website migration is valuable, but it is not a guarantee that organic...
Role-Based Access vs Blanket Bans for AI Coding Agents: What Goldman Sachs Shows
When an AI coding agent can read files, run commands, and change code, deciding who gets access is...
Google AI Classroom Part 2: Zero-Egress Browser Edge AI Investigator — Running Gemma 4 E2B, MediaPipe 3D Face Mesh & LiteRT Whisper in Web Workers
Author: Cyrus Wong (黃俊彥) — Google Developer Expert - Cloud (AI), Senior Lecturer at HKIIT ...
289 tests, 23 bugs, 8 real: what we learned checking every finding our AI testing agent made
Judge an AI testing tool by what survives a manual check, not by how many bugs it reports. I build...
29 checks said 0 subscribers. None of them clicked the link.
Twenty-nine checks recorded zero subscribers. The account had been disabled. None of the checks ever loaded the page a reader clicks.
How to Test an LLM When Pass/Fail Doesn't Work: A Practical Guide to AI Evaluation
A practical toolkit for evaluating LLM features: golden datasets, scored metrics, safety tests, CI gates, and production monitoring.
A caller said \"I'm his wife.\" Your AI changed the email, resent four invoices, and refunded $2,140.
A residential solar installer — 22,000 accounts across three states — found this one in a state...
Your LLM Capacity Plan Is Lying to You: Inside vLLM Overload Traces
During an overload experiment on vLLM, the engine's own prompt token counter missed a third to half...
Clean up messy product titles with one API call (free tier included)
If you've ever imported products from suppliers, marketplaces, or old spreadsheets, you know the...
Move your Vapi assistants to a self-hosted server with no platform fee
Hosted voice-agent platforms add a per-minute fee on top of what you pay your speech-to-text, model,...
Your agents are at home. You're at the office.
You kick off three agents on the Mac at home, then head to the office. Somewhere on the way one of...
Journey Before Destination: What The Stormlight Archive Teaches Us About AI Engineering
Why Soulcast pull requests and zero-shot roadmaps collapse when requirements shift—and why the engineering journey still matters.
Behavioral vaccination: a 400-row mixed prime protects a behavior
A 300 SST-2 + 100 AG News LoRA prime, merged with the AG adapter: sentiment 50->74%, AG News 45.8->71.2%, both p
Best of AI now speaks Spanish
Best of AI has been English-only since day one. That always felt like a strange limit for a directory...
My AI Tests Passed While Half the Mutants Went Undetected
If someone asked me how I verified code written by AI, I used to have a simple answer: I ran the...
Five Months Without Writing a Line of Code: The Basic Loop of Agentic Coding
How I built a JavaScript visualizer by instructing Claude Code in plain language, including what I got wrong, how I checked the work, and where this approach falls short.
Hybrid signatures have a trap: one valid half is not enough
Strip one signer from a COSE_Sign envelope and the other still verifies. How a post-quantum...