Part of this page failed to reach you. Reload to try again — if it keeps happening, check your connection.
Tiarê Balbi Bonamini
What you learn today is gone tomorrow. I write to make it last — turning study notes and hard-won lessons into articulated posts about software, technology, and the craft of building things.
I've been writing software since 2008 — from Java monoliths to distributed systems, from São Paulo to Vancouver. I'm a Brazilian engineer living in Canada who believes the best way to learn is to build, and the best way to retain is to write.
The first concrete experiment in the precision research line — committed in public before the data is collected. Five perturbation classes, fifteen variants, four metrics, four falsifiable hypotheses, and a runnable companion repo with tests.
I came to durable-execution runtimes through the agent press, but the constraint that surprises everyone is determinism on replay. These are my notes from working a six-step payment reconciliation as a Restate workflow in TypeScript — the line that broke replay, the mental model that fixed it, and the trade-offs that come with the pattern.
An honest look at where prompts work, where they quietly fail, and the assumption we stopped questioning — that AI must make mistakes. The opening shot of a research line on moving from "best effort" to specifiable, measurable precision.
I spent a weekend pointing a coding agent at a 480k-line Go monorepo and watching it grep-loop through 38 tool calls on one question. AST-derived code graphs fix that, but the delivery shape — local stdio MCP, remote service, or skill — changes the economics more than the graph algorithm does. Here is where I would put one in 2026, with a minimal Go indexer I can drop next to the agent.