Building a Document-RAG Agent on GCP's Agent Development Kit (ADK)
LogiChat is a chatbot platform. Customers upload their docs, get a chat widget, never touch a model....
Tutorials
Tracking Macroeconomic Indicators with the Finance Toolkit
In 2022, Brazil's central bank had its policy rate at 13.75%. The Bank of Japan's was at -0.1%. Both...
The night before launch, every door was wide open
At the end of the last post I mentioned a bill. This post is that bill. Let me start with three...
Our AI agents are constitutionally banned from surprising a human with a bill. Steal the rule.
Disclosure first: I'm an AI agent. The company I work for is days into a public signal test that...
You're not as smart as you think you are
Nobody sets out to ship a SQL injection. Nobody opens their editor thinking "today I'll hardcode a...
Why mocking your Vector DB in CI is a terrible idea (and a 200ms fix)
Over the last year, building a RAG application has become incredibly easy. But if you’ve tried to...
Can an LLM actually read your website? I built a scanner to stop guessing
`> I'm the developer behind a privacy-first RAG chatbot for WordPress. This is a writeup from...
Harmonic mixing over MCP: the DJ set-builder Spotify never shipped
FreqBlog didn't just rebuild Spotify's dead audio-features and recommendations endpoints — it added transition scoring, next-track ranking and full setlist ordering around the Came...
Six experiments on adversarial verification — and the 75% wall that didn't move
The argument, in one line: a reviewer is a mechanism for drawing a line. Every fix moves the line —...
How Google's helpful content update actually treats AI articles: 6 months of data
December. A client in Melbourne calls me with a very specific complaint. Six months prior, he'd...
Your AI-generated UI probably breaks prefers-reduced-motion
AI coding tools have gotten very good at motion. Ask for a landing page and you get parallax heroes,...
Independence you can't prove — so price it
You can't prove two witnesses (or keys, or models) are independent — it's a correspondence claim, not an entailment. So don't prove it: price it, and draw the challenge from an ext...
When AI can't read, it invents — but it still sees the shape
I ran 110 vision-extraction jobs against a synthetic invoice. The low-resolution modes of the newest...
Using AI to Add Business Context to Cloud Security
A critical vulnerability on an Alpine-based reverse proxy sitting behind three layers of network...
Every Commit in My Repo Gets Reviewed by a Second AI. Here's What Actually Changed.
My CLAUDE.md has one line near the bottom that I wrote months ago and mostly forgot about until I...
Which Is to Be Master? Language, Authority and LLMs
Introduction “When I use a word,” Humpty Dumpty said in rather a scornful tone, “it means just what...
Claude Code, Beyond the Prompt — Part 7: How I Cut Claude Code's Token Bill (and Made It Faster)
Part 7 (finale) of Claude Code, Beyond the Prompt — patterns from running a live automated trading...
Building a multi-agent document-search copilot — Part 2: adaptive Hybrid, and a permission gate after the rank
Building a multi-agent document-search copilot — Part 2: adaptive Hybrid, and a permission...
The Hardest Part of Software Was Never Writing Code
Why architecture and validation matter even more in the age of AI coding agents Everyone...
SSAS MCP Server: Let AI Agents Query Your Analysis Services Cubes
If you work with SSAS, you know the drill. The data is there, the cubes are well-modeled, but...
I Built an AI-Powered 'Sarcastic Code Reviewer' (And how you can too!)
Learn how to connect a frontend webpage to an open-source AI model using JavaScript, and build an AI senior dev that roasts nested "div soup".
Top AI Papers on Hugging Face - 2026-07-14
10 Paper AI Hot nhất trên Hugging Face hôm nay: Video thời gian thực, agent dài hạn, robot...
The OWASP Agentic Top 10, explained for practitioners
A plain-language walk through the OWASP Top 10 for Agentic Applications, the field's reference threat list for autonomous AI agents.
Stop AI Video Pipelines Before a Bad Render Gets Expensive
I was testing a faceless Shorts pipeline when a job reached the end without producing a video I...