I’m Antão Almada, a software architect focused on how AI is reshaping the software development lifecycle. My work centers on coding agents, deterministic workflows, and the engineering practices required to make AI a reliable part of real systems — observability, governance, reproducibility, and domain‑knowledge preservation.

I lead architecture initiatives at NTT Data Portugal, helping teams adopt GenAI‑driven development, integrate GitHub Copilot, and design agent‑based workflows that scale across projects. My goal is always the same: bring clarity, determinism, and engineering discipline to a space that often defaults to improvisation.

Before focusing on AI, I worked across .NET backend architecture, IoT, XR, and distributed systems. My long journey with .NET — which I wrote about in the My adventure with .NET development post — still shapes how I think about boundaries, contracts, and domain behavior.

A major influence early in my career was Scott Meyers’ Effective C++ series. Those books taught me that mastering a programming language is not enough — you must understand what the compiler does with your code and how the processor executes it. That mindset still drives my interest in performance, which I explored in the Performance optimizations post.

I’m also active in open‑source, particularly through the NetFabric organization. NetFabric is where I explore ideas around modern C#, high‑performance data structures, functional patterns, and domain‑driven abstractions. These projects often influence the patterns I write about here, especially when bridging deterministic code with non‑deterministic AI.

Outside of work, I’ve always had a passion for bike riding, especially BMX — the photo on this page is a reminder that I like building things that move fast, whether they’re systems or bikes.

I’m also passionate about photography. Every picture on this blog is original and taken by me. If you want to see more of my work, you can check my Instagram: https://www.instagram.com/antao.almada/

This blog is where I document the engineering side of AI in software development:

• how coding agents behave in real workflows
• how to govern and constrain them
• how to reduce token usage and improve efficiency
• how to preserve domain knowledge during modernization
• how to design hybrid workflows that mix code and AI
• how to build reproducible, tool‑agnostic engineering practices

If you’re working on similar problems — AI engineering, modernization, or open‑source tooling — you can find me on LinkedIn or explore my projects on GitHub. I’m always open to good conversations and new ideas.