Sign in to view Jakub’s full profile
or
New to LinkedIn? Join now
By clicking Continue to join or sign in, you agree to LinkedIn’s User Agreement, Privacy Policy, and Cookie Policy.
Sign in to view Jakub’s full profile
or
New to LinkedIn? Join now
By clicking Continue to join or sign in, you agree to LinkedIn’s User Agreement, Privacy Policy, and Cookie Policy.
Greater Munich Metropolitan Area
Sign in to view Jakub’s full profile
or
New to LinkedIn? Join now
By clicking Continue to join or sign in, you agree to LinkedIn’s User Agreement, Privacy Policy, and Cookie Policy.
256 followers
256 connections
Sign in to view Jakub’s full profile
or
New to LinkedIn? Join now
By clicking Continue to join or sign in, you agree to LinkedIn’s User Agreement, Privacy Policy, and Cookie Policy.
View mutual connections with Jakub
or
New to LinkedIn? Join now
By clicking Continue to join or sign in, you agree to LinkedIn’s User Agreement, Privacy Policy, and Cookie Policy.
View mutual connections with Jakub
or
New to LinkedIn? Join now
By clicking Continue to join or sign in, you agree to LinkedIn’s User Agreement, Privacy Policy, and Cookie Policy.
Sign in to view Jakub’s full profile
or
New to LinkedIn? Join now
By clicking Continue to join or sign in, you agree to LinkedIn’s User Agreement, Privacy Policy, and Cookie Policy.
View Jakub’s full profile
-
See who you know in common
-
Get introduced
-
Contact Jakub directly
Other similar profiles
Explore more posts
-
Marco Möller
Pionix GmbH • 14K followers
I just read about an interesting - but frankly, scary - shift happening in the "wild" of the software ecosystem. It’s a thought that has been haunting open-source maintainers lately, articulated sharply by Sascha Pallenberg 潘賞世 regarding situation in the "Tailwind CSS open source ecosystem." The premise: We are witnessing the onset of "AI cannibalism" for Open Source. The mechanism is simple: As AI becomes our primary interface for coding, we stop visiting documentation. We stop hanging out in GitHub Discussions. We stop engaging with the humans behind the code. The "Source" stays open, but the "Community" gets dissolved and its sources of funding dry up. AI Cannibalism Loop: This leads to a paradox where AI is effectively cannibalizing its own future. LLMs are trained on the public artifacts of human collaboration—the forum debates, the PR comments, the documentation fixes. By providing "good enough" answers in private chats, AI short-circuits the very collaboration that creates that data. • If you never visit the docs, you never spot the typo. • If you never struggle through a bug in a forum, you never submit the Pull Request. • If the Pull Request doesn't exist, the next model has nothing new to learn from. We are eating our seed corn. We are using AI to harvest the fruits of open source while simultaneously poisoning the soil that grows it. Firmware / industrial tech (like the EVerest Project) has a "Hardware Shield": In my opinion, projects like EVerest are slightly more insulated from this atrophy, but only by necessity. When you have a massive "zoo" of variable hardware in the loop and the stakes of a bug include literal smoke or fried electronics, you cannot afford to be a passive consumer. The physical reality of a charging station forces you back into the loop. Technoeconomic Reality: Open source isn't a "free lunch" ⮕ it is a shared R&D department. Proprietary software is expensive because of licensing. Open source is "free" because of the contribution flywheel and paid premium products. If that flywheel stops because we have all retreated into our private AI silos, the TCO (Total Cost of Ownership) will eventually skyrocket as projects become unmaintained, insecure, and "stale." Are you seeing a drop in community engagement in your projects since the LLM boom? Is the efficiency gain of the "private chat" worth the risk of project decay?
20
6 Comments -
InfoQ
30K followers
Software architecture is a financial option. 📈 The more uncertainty a project faces - the more valuable architectural flexibility becomes. Gregor Hohpe explains why this isn't just a metaphor - it's backed by mathematics. 🔗 Explore the full talk [transcript included]: https://bit.ly/40sdvP0 #SoftwareArchitecture #InfoQ
11
5 Comments -
FrontKon - Prague's Frontend Conference
214 followers
One wrong Service Worker update can lock your users on a broken version forever. Here is how to do it right. 🚨 Caching data at the client edge is the ultimate way to boost frontend performance, but managing offline storage at scale requires serious architectural precision. Catch Dominik Dvořák (Senior Frontend Developer at Livesport) live on our Partner Stage at #FrontKon26! 💡 What you will learn: • Battle-tested caching patterns for high-concurrency web apps. • How to safely invalidate and version Service Worker caches. • Implementing a reliable killswitch when production incidents strike. Stop wasting bandwidth and start caching like a pro. 📍 Partner Stage | 🇨🇿 Presented in Czech 🎟️ Join us at FrontKon in October. Secure your spot today: www.frontkon.tech 🙌 Special thanks to Livesport for their support as our conference partner!
6
-
Eficode
21K followers
#GitHubUniverse 2026 is almost here! And with it comes exciting developments in AI, developer productivity, and automation. But once the announcements are over, the big question is: Which updates matter most for your team, and how can you put them into practice? Join our GitHub expert Maik Müller for a post-Universe webinar. We'll go beyond the announcements to explore what the latest innovations mean for your team and how to turn them into real results. Here's what we'll cover: ✅ Key announcements: The biggest updates from GitHub Universe 2026. ✅ Business impact: What they mean for developer experience (#DevEx), security, and faster releases. ✅ Practical adoption: How to introduce new capabilities without disrupting your existing workflows. ✅ Live Q&A: Get your questions answered by our GitHub expert. November 10, 2026. | 3:00 PM GMT Ready to turn GitHub's latest innovations into real business value? Register for free https://efcd.co/4z0kvB9
9
-
Atlanta Java Users Group
2K followers
Slow database queries killing your Java application’s performance? 🐢 At #Devnexus, Markus Kett dives into how modern caching strategies can dramatically improve application speed and responsiveness—without putting unnecessary load on your database. 🔗 https://lnkd.in/eGXxiRZC In this session, you’ll learn why traditional database-driven architectures often become performance bottlenecks, how caching can help overcome those limits, and how to choose the right approach—whether in-memory, distributed, or hybrid—for your Java applications. Walk away with practical insights you can apply to build faster, more scalable systems. 🎟️ Get tickets at devnexus.com 📨 Stay up to date: https://lnkd.in/eNQVASpk
2
-
Game of Codes - IT conference Niš
999 followers
How fast can you make a bad Spring Boot application? Vlad Mihalcea is taking the worst possible Spring Boot application and optimizing it LIVE on stage - with and without AI. No slides-only theory. Expect live coding, profiling and performance tuning. Using tools like VisualVM, IntelliJ IDEA Profiling and Hypersistence Optimizer, Vlad will show how to identify bottlenecks, optimize data access and build a batch processing task capable of inserting data up to 10× faster. And there’s an extra question behind the whole session: 🤖 How much can AI actually help when performance really matters? 📍 Game of Codes 2026 📅 October 10 · Niš Get ready for some serious Spring Boot performance tuning. 🔥
6
2 Comments
Explore top content on LinkedIn
Find curated posts and insights for relevant topics all in one place.
View top content