Alexandre Borges

Many software teams choose to automate their returning (administrative) tasks. Why? Less waiting, less errors, more focus. The result: better quality in less time! But how do you do that? 9 out of 10 times, a simple command-line tool does the trick. How to build that? - A script? 🤔 Just not powerful enough. - Go or Rust? 🛠️ Don't know these languages yet... - Java (or C#)? ☕ But it should start fast and work fast! In my latest blog, Building a CLI with Quarkus, Kotlin and GraalVM, I explain how you can get get started to build such tools using a modern, JVM-based stack. Check it out and let me know: which task in your team is the first one you'd want to automate? ⤵️ https://lnkd.in/eAwKr56t

23

𝐉𝐚𝐯𝐚 25 𝐨𝐧 𝐀𝐖𝐒 𝐋𝐚𝐦𝐛𝐝𝐚 🚀 Cool news for Java folks! AWS Lambda now supports Java 25 and it already takes advantage of some of the start-up performance work happening in Project Leyden. 📄 Blog post: https://lnkd.in/exDZuvcQ Java 25 brings AOT caches by default, giving faster cold starts and lower memory usage compared to the old CDS approach. Great to see big cloud vendors adopting new JVM improvements 🎉 ⚠️ 𝐍𝐨𝐭𝐞: Java 25 runtime isn’t available in all AWS regions yet — for example, it’s still missing in eu-central-1. Also, great talk on Project Leyden by Moritz Halbritter and Ana-Maria Mihalceanu 👇 https://lnkd.in/eWmNvfsQ #Java25 #AWSLambda #ProjectLeyden

149

9 Comments

What a second day at Azure Dev Summit 2025 in Lisbon My main takeaways: • Scott Hanselman and Scott Hunter: The next 25 years of software engineering. AI won’t replace developers, it will augment them. • Maarten Balliauw: HTMX and ASP.NET Core. A simpler path to interactivity without heavy front-end frameworks. • James Montemagno: Building secure and scalable MCP servers. Model Context Protocol connects agents and APIs safely. • Chris Ayers: The Well-Architected Architect. Make trade-offs explicit and measurable across cost, reliability, security, and performance. • Andrew Westgarth: .NET Aspire for AI apps. Easier orchestration, observability, and AI integration in modern .NET. My reflection: • MCP and .NET Aspire are shaping how we assemble architectures faster and take prototypes to production. • The center of gravity is shifting from writing code to composing intelligent, observable systems. • Architecture is back at the core of software engineering. #AzureDevSummit #AgenticAI #Azure #DotNet #MCP #Aspire #Architecture #Microsoft

20

2 Comments

Developers spend more time reading code than writing it—so why aren’t we trained for it? Hannes Lowette & ☕️🥑 Marit van Dijk van Dijk discuss strategies for mastering code comprehension, from structured exercises to leveraging IDE features.  https://lnkd.in/dMA77PqV

20

It is super awesome to see the push towards the zero-cost abstractions in .NET. We all know about spans, but there specifically designed for high performance and require the code changes to migrate to them. But now, due to de-abstraction feature in .NET 10 we can eliminate all the overhead in the existing cases. For instance, if you have Enumerable.Any with a delegate that captures instance state, the runtime can stack-alloc it, and (if it's small) to inline it. This will achieve the performance of a manual code that obtains span out of a list and "inlines" the delegate logic into the loop itself! More info on that? Here: https://lnkd.in/ejpa3546 And as always, thanks for sharing this post!

45

Continuous Integration: is a software development practice where each member of a team merges their changes into a codebase together with their colleagues changes at least daily. Each of these integrations is verified by an automated build (including test) to detect integration errors as quickly as possible. Teams find that this approach: - Reduces the risk of delivery delays, - Reduces the effort of integration, and - Enables practices that foster a healthy codebase for rapid enhancement with new features. Here are key practices of continuous integration from Martin Fowler:

14

2 Comments

Hey junior and medior developers. Here is my strategy for tackling any use case / feature and turn it into code: Before you even type or generate a single piece of code, you should ask yourself some questions. First thing to know: What are the desired outcomes / results of this feature? You need an understanding of what is expected from the process you are going to code. If the process should return data, what are the fields that need to be returned? In which format? Second: What will be the input of this use case or feature? You also need to understand the start situation of the feature. What information is available and in which form/format? This information can be the foundation of a first test:   Arrange - > Set up the start situation you just learned about Act - > Trigger the process Assert - > The expected results you just discovered. You are not done yet. In order to break down this feature into code components, we need to understand the process a bit more. What are the different steps in this process? A step is a collection of operations that together perform a meaningful task. This is you figuring out how you will go from the start situation to the desired results. In your ‘blueprint’ design, each one of these steps is a code component. This is a high level overview of how I tackle use cases / features. You want a detailed example of how this works? I am working on a course that will explain all this in detail. Follow me to stay updated on that! #softwaredevelopment #softwareengineering #softwaredesign

10

5 Comments

👨🏻‍💻 Jaco van Gelder just published a great article about getting started with Lakebase! Check the link in the comments, and here is a short overview of practical lessons: 🧱 Start with a clear understanding of where Lakebase fits in your architecture. It's easy to think of it as just another service to plug in, but it actually plays a much bigger role. It adds an operational layer on top of your lakehouse that changes how you design and build your data products. 🧱 Treat your Python environment as part of your architecture. Using modern tooling like "uv" makes your setup reproducible and production-ready from the start. 🧱 Keep deployments metadata-driven. Leaning on YAML configurations makes your setups cleaner, easier to version and simpler to automate. 🧱 Think beyond experimentation: structure your project as if it will go to production (because it probably will). Dtyped Databricks #databricks #lakebase #python #uv

95

3 Comments

We released the 𝗹𝗮𝘀𝘁 #AxonFramework 4 minor version, version 4.12.0, last Friday. It's been quite a journey since the first AF4 release, which we made in March 2019(!). We introduced many worthwhile features, like reset handlers, exception handlers, and annotated interceptors. New, more efficient ways to process events, support for combining different event sources, and dead-letter queues. What about thorough tracing support and micrometer monitoring? And the notion of an aggregate's creation policy, or the option to construct a polymorphic aggregate. So, what does 4.12.0 add to that list? Not a ton, to be entirely frank with you. We view this minor release as a stopgap before fully transitioning our focus to AF5. As such, it contains a set of internal dependency upgrades, a topology change listener, a simplification of the authorization interceptors, and several documentation fixes. I expect a patch release in the near future. I mean, there's never been a release without a patch release (I guess we're not perfect either 🤷). This will either be released due to some bug on our end or a newly uncovered CVE in a dependency AF uses. To top off AF4.12, we will make a "deprecation release" right about when we release AF5. As that release literally just adds deprecation warnings and documentation, I argue it's fair to make that a patch release —a patch release to help everybody during the migration to AF5. (Thanks to all who voted on my poll about this, giving me confidence that this is the right way to go.) Although I say this quite frequently, I cannot stop doing so: thanks to everybody who's contributed to AF4 in one way or another! Whether it was a simple comment, an issue, or a pull request you added. Whether you were a part of @AxonIQ or somebody who enjoys using this piece of open-source software for the solutions it provides. The 𝗹𝗮𝘀𝘁 𝘀𝗶𝘅 𝘆𝗲𝗮𝗿𝘀 wouldn't have been the same without all that, so thank you!

34

2 Comments