🔷C# Solid Principles The five solid principles are good for Object-Oriented design. It's always recommended to keep these principles in mind when writing new code. 🧵Thread

Today an Amazing day at Milan with @gdgmilano. AI speeds up the answer but it's the people we meet at events who truly shift out perspectives. The value of human connection remains unique and irreplaceable.

Another explanation for managers shipping more: They see the entire backlog. AI lets them finally tackle all those "someone should fix this someday" tasks without stealing sprint capacity from the team. Interesting shift.

One fascinating data point from LeadDev: Engineering managers using AI reportedly ship around 4× more code than before. Why? Maybe because managing AI agents looks surprisingly similar to managing people: delegate, review, iterate.

A question we kept coming back to: Where does quality actually live? Is QA a gate at the end? Or is quality something owned continuously by the whole team? AI makes this question much harder—and much more important.

AI can make you write code faster But... What if coding isn't your bottleneck? If your delivery pipeline spends weeks in reviews, QA, approvals and deployments, 10x faster coding may translate into almost no improvement for users Optimize the constraint, not the keyboard

One mindset shift stood out. For years, good engineering advice was: Don't chase every new tool. With AI, that advice is changing. If you don't regularly experiment, your mental model of what's possible can become outdated surprisingly fast.

Architecture is not about drawing diagrams. It's about making decisions that keep the system adaptable, maintainable, and ready for change. Sometimes the best decision is: "not yet."

How is AI actually changing the way we build software? The most interesting discussions weren't about technology. They were about bottlenecks, quality, trust, and how engineering itself is evolving

The cost of changing a decision matters more than the decision itself. Smart architects focus on making choices reversible whenever possible.

When requirements are still evolving, flexibility is a feature. Locking into a technology too early can be more expensive than delaying the decision and learning a little longer.

Architecture should optimize for learning, not certainty. Every iteration gives you better data. Every experiment reduces assumptions. Every postponed commitment improves decision quality.

Most architectural mistakes don't come from bad decisions. They come from making reasonable decisions too early, with incomplete information.

Good software architecture creates options. Great software architecture preserves options. The ability to defer a decision is often more valuable than making the "right" decision early.

The goal of architecture isn't to predict the future. It's to make future changes affordable. Systems that evolve well are often designed by teams that resist the urge to decide too soon.

A useful question for architects: "Do we really need to decide this today?" If the answer is no, preserving optionality may be more valuable than choosing a solution.

Experiences like this are a reminder of why working in AI is such an exciting journey Real challenges, great minds, ideas becoming real projects, and one clear takeaway: true innovation happens through collaboration Bootcamp wrapped up — energy high, inspiration even higher🚀🤖

Every architectural decision is a trade-off. The earlier you make it, the less information you have. The later you make it, the more confidence you gain. Architecture is not about certainty. It's about managing uncertainty.

Not every tech event needs a conference room. Yesterday at #TalkAndWalk2026 we proved that you can talk about technology, share ideas, and build meaningful connections… simply by walking together. 🌿🚶‍♂️⚡

Premature optimization isn't just a coding problem. It's an architecture problem. Choosing databases, frameworks, messaging systems, or microservices too early often creates constraints before you understand the real problem. Delay commitment. Learn first.

A surprising software architecture principle: ❌ Decide everything upfront ✅ Decide at the last responsible moment The more uncertainty you have, the more valuable flexibility becomes. Good architecture buys you time.