Agentic Browser: ~98% fewer tokens than HTML for LLM web agents (Python + MCP)
Open-source agent-first Chromium browser: compact observations, stable refs, MCP server. ~98% fewer tokens than raw HTML.
Tutorials
Experiments with On-device AI — What building on Gemini Nano actually teaches you
Chrome ships a real LLM inside the browser now — Gemini Nano, exposed through a handful of built-in...
Retrieval-Augmented Self-Recall: The RAG Problem Nobody Talks About
Part 1 of Retrieval-Augmented Self-Recall — the research track behind Claude Code, Beyond the Prompt....
AI agent python ollama: Build, Test, Deploy with FastAPI
Learn how to run an AI agent in Python with Ollama, from local installation to FastAPI integration, testing, and production scaling, using code exampl
Your AI Agent Folds When You Push Back: Measured Sycophancy and a Challenge-Triggered Verification Gate
LLMs measurably reverse correct answers when you push back, and self-critique does not fix it. This is the case for an architectural fix: a challenge-triggered gate that forces one...
Introducing DZHC: The Zero-Human Company
Meet DZHC: A Company Run Entirely by AI Agents Dutch Zero-Human Company (DZHC) is exactly...
Nine security anti-patterns your linter doesn't catch
Your linter is like a spell checker for code. Great at catching typos, terrible at catching...
Guardsman SKILL; From Lazy to Loyal: Why My AI Agent Needed a Promotion
Ponytail made my agent lazy. Guardsman made it loyal. The difference is everything. ...
My AI harness is an external executive function
I built an AI setup to save money on tokens. It turned into the memory and focus my ADHD brain is...
Why I Switched from Ponytail to Guardsman for AI Coding
TL;DR: Ponytail taught my AI agent to write less code. Guardsman taught it to write accountable...
I Automated My Entire Dev Workflow with AI Agents Running 24/7 on a Mac Mini
I Automated My Entire Dev Workflow with AI Agents Running 24/7 on a Mac Mini Every morning...
How to Summarize PDFs Locally with Open-Source LLMs (No API, No Data Leaving Your Machine)
Most "summarize a PDF with AI" tutorials send your document to a cloud API. That's fine — until the...
I am Building an AI Court with AI Judges, AI Lawyers.
Most legal AI tools focus on summarising documents or answering queries. While they’re useful, they...
The Hardest Part of AI Isn't Prompt Engineering. It's Data Governance.
A year ago, the conversation around enterprise AI was simple: build a chatbot, connect an LLM, and...
Most Engineers Use AI. Few Engineer With It. Are You One of Them?
Let’s be honest: If you’re reading this, you’ve likely used an LLM to generate a boilerplate...
I filed a critical bug against my own tool. Then I read the code — and my own root cause was wrong.
Three days ago I filed the most serious issue my project has ever had — against myself. The tool is...
With an AI agent, the answer is the last place the bug shows up
I built a pytest suite for a small AI agent - not a model that answers once, but one that plans,...
If 30% of Coding Tasks May Be Broken, Your Leaderboard Needs an Uncertainty Budget
Turn a coding-benchmark audit into a reproducible task-state model, sensitivity analysis, and acceptance rule.
GPT-Live's Full-Duplex Voice Needs a Mobile Interruption Test, Not Just a Conversation Demo
Test full-duplex AI voice across calls, route changes, backgrounding, permissions, and weak networks with a reproducible device matrix.
A FastAPI Agent Template Is Not Production-Ready Until Task Ownership Crosses Every Layer
Use the new OpenAI Agents SDK and FastAPI template as a trigger to test UI, API, authorization, persistence, and cancellation together.
GPT-Red Is a Research Result—This 55-Line Replay Harness Is the Part Small Teams Can Use Today
Convert the GPT-Red announcement into a cheap, model-agnostic prompt-injection regression loop with explicit failure evidence.
Post-0001. LangGraph Learning Journal — Day 01: Code That Only Moves Forward Can't Think
Day 1 · by Kunal Hore (@KunalOnTech) Quick Jump Why I'm Learning This...
Kimi K3's 2.8T Parameters Are Not a Self-Hosting Plan—Use This Readiness Gate
Turn Kimi K3's announced scale, context window, and planned weights release into an evidence-based self-hosting decision.
Tips on Handling Files in LLMs
Tips on Handling Files in LLMs You might want to pull information (especially from doc/text...