https://aalmada.github.io/The official blog of Antão Almada. A blog about software development and technology. 2026-06-11T22:02:01+01:00 Antão Almada https://aalmada.github.io/ Jekyll © 2026 Antão Almada /assets/img/favicons/favicon.ico /assets/img/favicons/favicon-96x96.png 2026-06-10T00:00:00+01:00 2026-06-10T23:20:46+01:00 https://aalmada.github.io/posts/Copilot-is-a-Ecosystem/ Antão Almada

For a long time, “using GitHub Copilot” meant typing a prompt in VS Code. That was the Copilot we knew: the one embedded in the editor, the one that completed code, the one that felt like a natural extension of the IDE. But that Copilot is only one of several. GitHub Copilot appears in multiple places, with different capabilities, different interaction models, and different agents behind them, ...

2026-06-08T00:00:00+01:00 2026-06-08T00:00:00+01:00 https://aalmada.github.io/posts/Agent-Governance/ Antão Almada

Coding agents have reached a point where they can perform meaningful, multi‑step work across an entire codebase. They can refactor dozens of files, rewrite architectural boundaries, introduce or remove dependencies, modify infrastructure code, run shell commands, and apply patches at scale. Once an agent can do all of that, we no longer have a convenience feature. We have automation with conseq...

2026-06-07T00:00:00+01:00 2026-06-08T22:49:33+01:00 https://aalmada.github.io/posts/Hybrid-workflows/ Antão Almada

Deterministic vs Non-Deterministic Work GenAI is extraordinary at handling non‑deterministic workflows: the kind of work where ambiguity exists, where interpretation matters, and where the “right” answer depends on context rather than rigid rules. These are the domains where reasoning models shine: summarizing, rewriting, classifying, interpreting, and making sense of messy human inputs. But ...

2026-06-06T00:00:00+01:00 2026-06-08T22:49:33+01:00 https://aalmada.github.io/posts/Observability-for-coding-agents/ Antão Almada

Coding agents were fun when they felt like magic. You typed a request, something happened behind the curtain, and results appeared. Then usage‑based billing became the norm. The curtain stopped being charming and started being expensive. In my previous post, Reducing Token Usage in Code Agents, I focused on shrinking prompts, improving caching, and using semantic context to avoid unnecessary mo...

2026-06-06T00:00:00+01:00 2026-06-08T22:49:33+01:00 https://aalmada.github.io/posts/From-Grep-to-Graph/ Antão Almada

We have spent the last few months watching coding agents evolve from clever autocomplete to something closer to a real engineering component. We now package them, version them, sign them, distribute them, and monitor them. We treat them as artifacts with lifecycles, not prompts floating in the void. But there is still a missing layer in this stack: the layer that determines whether an agent tru...