If you've tried to follow any AI coding discussion in the last six months, you've probably felt like everyone suddenly started speaking a dialect you never signed up to learn. "Vibe coding." "Agentic workflows." "Context windows." "Prompt engineering." The jargon is multiplying faster than JavaScript frameworks, and that's saying something. Matt Pocock — who you might know from his TypeScript educ
GitHub Copilot just got a lot more complicated — and not in a good way. If you tried to sign up for Copilot Pro recently and hit a wall, that's not a bug. GitHub quietly paused new sign-ups for Copilot Pro, Pro+, and Student plans starting in late April 2026. No end date announced. No workaround offered. Just a message and a door that won't open. That alone would be worth covering. But they made t
Book: TypeScript in Production Also by me: The TypeScript Library — the 5-book collection My project: Hermes IDE | GitHub — an IDE for developers who ship with Claude Code and other AI coding tools Me: xgabriel.com | GitHub You have seen the shape of this incident before. A 500 lands in production. The frontend says "checkout failed". The Hono service that owns /checkout called the prici
Every observability vendor has bolted "AI" to their landing page. Half of those features are genuine improvements. The other half are autocomplete in a costume. After a few years of running these tools across enterprise estates, here is where AI-augmented SRE actually pays off, where it doesn't, and what we'd advise teams adopting it today. The single most defensible use case. A medium-sized estat
Anthropic now ships at least three different memory models inside the Claude product family, and they don't behave the same way. Claude.ai has a chat memory feature for Pro, Max, Team, and Enterprise users that summarizes prior conversations and injects that summary into new chats. Claude Code has CLAUDE.md files plus a separate "auto memory" directory the model writes to itself, both loaded at se
Iris v0.4.0 ships today. It's the release where protocol-native eval crosses from "deterministic rules" into "semantic scoring" — without giving up any of what made the deterministic layer work. Three headline features plus a lot of infrastructure work that quietly compounds. I'll go through each, why it matters, and how it fits the thesis. Heuristic rules catch a lot: length, keyword overlap, PII