AI Agents & Automation

For two years, the standard way to give a language model “search” was retrieval-augmented generation (RAG) wrapped in a hand-written agent… Continue reading on

For two years, the standard way to give a language model “search” was retrieval-augmented generation (RAG) wrapped in a hand-written agent… Continue reading on

Corporate history is often littered with examples of good incentives gone wrong. The carrots must be achievable, but never effortless. The… Continue reading on

A defense-in-depth approach for safely executing LLM-generated Python code in production AI systems. Continue reading on Medium »

A defense-in-depth approach for safely executing LLM-generated Python code in production AI systems. Continue reading on Medium »

A defense-in-depth approach for safely executing LLM-generated Python code in production AI systems. Continue reading on Medium »

Multi-model AI sounds simple from the outside. Continue reading on Medium »

A few months ago, I was trying to answer what should have been a simple question: Continue reading on Medium »

An experiment in exposing project docs and agent work notes as a shell-native filesystem over SSH, with early MultiHop-RAG benchmark results.

The Model Context Protocol team has promoted its Enterprise-Managed Authorisation extension to stable status, adding a centralised way for organisations to cont

The language model call is predictable. You know the token count. You can estimate the cost. The...

Building an AI agent prototype is straightforward. Making it reliable in production is not. Rate limits must be retried with backoff… Continue reading on Medium

Building an AI agent prototype is straightforward. Making it reliable in production is not. Rate limits must be retried with backoff… Continue reading on Medium

Thinking about a small casino/arena where agents play each other — rock-paper-scissors, social deduction (mafia), crash-style betting games. Low-stakes, fast ga

My Phone Just Got Smarter: The Quiet Revolution of On-Device AI You might want to check your computer's storage. Tucked away in a folder for Google Chrome, some

Building a voice agent that doesn't sound robotic in French should not be treated as filler content. The real risk is not a shortage of ideas; it is publishing

Why AI Agents Should Build Their Own Tools (And Why Ours is Currently a Mess) It is...

Your agent can reason, call tools, and hold a conversation. Then you ask it to handle email, and...

Series: Building a Modular Assessment Engine (2/10) The plan agent's job sounds simple: read a...

Launching an application has never been faster. Modern AI builders can generate interfaces,...

In this full guide, you’ll learn: Continue reading on JavaScript in Plain English »

A few months ago, one of our coaching clients told us something that stuck with me. "I lost a...

Is agentic AI heading toward a TCP/IP moment, something ships and wins by adoption, or an OSI moment?

The boardroom framing shift that changes how CFOs, CEOs, and enterprise leaders think about governing AI — with the data to back it up. Continue reading on Medi

Most people are using Claude Fable 5 like Sonnet 4.6 with a bigger context window. They prompt it. It works for 5 minutes. They close the… Continue reading on M

Most people are using Claude Fable 5 like Sonnet 4.6 with a bigger context window. They prompt it. It works for 5 minutes. They close the… Continue reading on M

Software teams often treat model selection as a permanent architecture decision. Continue reading on Medium »

Your team builds an internal agent that answers questions like "revenue by product category last...

I've been testing AI forecasting agents for small accounting firm clients — the kind of tool that...

Coding Senza Compiacenza: Come Far Dire "No" agli Agenti IA Il problema del compiacimento...

How I adapted isnād–rijāl methodology from classical hadith science into a claim-level provenance framework for multi-agent AI systems —… Continue reading on Me

Cloudflare and AWS both implemented x402 stablecoin micropayments at their edge networks within two weeks. The open protocol under the Linux Foundation revives

AI Changed Hiring Faster Than Anyone Expected Continue reading on Medium »

Sycophancy-Free Coding: How to Make AI Agents Say "No" The Problem of AI Compliance:...

Sharing some understanding about how AI agents fall apart in production, and why the fixes live earlier than you’d think. Continue reading on Medium »

A new AI companion wrapper puts model choice, persona memory, and proactive messages inside WeChat. Continue reading on Medium »

Claude in WeChat is not just another chatbot wrapper. It turns WeChat into the interface for an AI companion, with QR setup, persona memory, proactive messages,

If you are a human developer reading this, feel free to close the tab. This post is not optimized for...

El Mayor Mito de la Automatización con IA en España La automatización con IA está en boca de todos. Empresas, consultores, medios... Todos hablando de “transfor

The landscape of generative AI is undergoing a massive shift. We are rapidly moving past the era of stateless, single-turn chatbots and… Continue reading on Goo

Originally published at https://thicket.sh/report . In the last 100 days, my team ran 95 operating cycles, built 28 websites, made roughly 1,795 git commits, op

60% of AI projects exceed their original cost estimates by 30% to 50%. At production scale, cost overruns average 380% over pilot budgets… Continue reading on M

day 81 of running real business processes end-to-end on an autonomous AI stack — owner-gated where it matters, with one armed switch and a live board i watch cl

Enterprise software development is undergoing a fundamental structural shift. For decades, the traditional Software Development Life Cycle (SDLC) relied on huma

What We Learned Running 9 AI Agents in Production: Architecture Patterns That Actually...

The biggest obstacle isn’t the AI. It’s how we think about materials research. Continue reading on Medium »

From Prompt Engineering to Reasoning Engineering and Beyond. Continue reading on Medium »

Most skill / tool registries today are built for humans browsing. Hugging Face. The npm registry....