Technical articles on AI safety, agentic design, and building autonomous systems responsibly.
A breakdown of the architecture patterns, memory systems, and tooling I use to keep token consumption lean across multi-agent AI workflows. No model switching required.
I haven't used Kimi K2.5 myself, but I've done the research. Here's my honest take on Moonshot AI's cost play, when it makes sense, and why OpenRouter is the right way to test it.
A deep dive into OpenRouter, the unified API that gives you access to 200+ AI models through a single endpoint. How it works, what it costs, and when it makes sense over direct API access.
Part 3 of the CORE Discovery series. A hands-on walkthrough covering local setup, Azure configuration, and running your first discovery session through all four phases.
Part 2 of the CORE Discovery series. A technical deep-dive into the provider abstraction pattern, AI agent architecture, and design decisions behind a portable full-stack AI application.
Part 1 of a three-part series introducing CORE Discovery, an AI-powered platform that coaches product teams through structured customer discovery.
A practical guide to theGuard's CLI. How to turn protection on and off, add custom rules, manage your guard from the terminal.
How the adapter pattern lets theGuard protect six different AI coding tools through a single shared rule engine.
Why traditional verification fails when the AI agent can read the code, and how theGuard implements true two-factor authentication.
Page 1 of 9 (73 posts)
