What it's like when it gets you.

Real moments from building Chalie in the open — the day it remembered something we'd half-mentioned, the Skills that write themselves, and the lessons in between.

23 Jun 2026

Ten Models, One Assistant: Which Engine Drives Chalie Best?

We put ten language models — frontier and open-weight — through the same battery of everyday tasks inside Chalie, then scored each on accuracy, speed, cost and how cleanly it drove. The spread wasn't what we expected.

By Dylan Grech

30 May 2026

The Skill System That Writes Itself

Every agentic AI shipped 'skills' months ago. Chalie waited — and built a skill system that fills itself, cleans up after itself, finds the right one on its own, and quietly shapes each one to fit you.

By Dylan Grech

29 Apr 2026

The Day Chalie Scheduled a Reminder to Ask About My Doctor's Appointment

While testing v0.5.0, I mentioned a minor medical thing in passing. The next day I opened the scheduler and found Chalie had quietly set itself two reminders to follow up. A look at the three systems that combine into something that feels like caring.

By Dylan Grech

16 May 2026

Directing Agents, Not Writing Code: How Chalie Develops Itself

An inside look at the autonomous development workflow behind Chalie — where AI agents build features, run regression gates, and improve themselves nightly while a human steers direction.

By Dylan Grech

21 Apr 2026

The Meta-Harness, Two Weeks Later: From Interesting Experiment to Working Contributor

A follow-up on our self-improvement loop. We tore out most of the orchestration, made the kernel a single skill, and watched it open and merge multiple PRs in a single afternoon.

By Dylan Grech

6 Apr 2026

Building an AI That Improves Itself: The Meta-Harness

How we built an autonomous improvement loop that uses Claude Opus to analyze test failures, propose fixes, and create pull requests — with no human writing code.

By Dylan Grech