PastAGI

PastAGI is an AI engineering publication for builders who want the math, not the marketing. Every piece is grounded in something checkable: serving math, eval methodology, or a cost audit with real numbers. Recent work includes an audit of GPT-6 Astra pricing at $6 an hour, a teardown of how custom AI inference chips are taking share from GPUs, a guide to why AI agents leak sensitive data and the controls that stop them, prompt dependency graphs that shrink retest sets, and a review of the TontaubeV1 character-level TTS model with the serving math laid out. The editorial line is anti-hype: chain-of-thought transparency, detector false-positive rates, and benchmark skepticism get the same scrutiny as vendor claims. Categories cover guides, tools, engineering deep dives, and real use cases. If you ship AI features and have been burned by a demo that fell apart in production, this is written for you. Read it at https://pastagi.com/.