Heapcon

And just like that, Heapcon 2024 has come to an end! :D From the stunning design everyone’s been buzzing about to a lineup of talks that had something for everyone, this year’s conference felt truly special. We loved seeing old friends reconnect – from former colleagues to professors and teammates from past companies – all coming together under one roof. That’s the magic of Heapcon! Huge thanks to Madlenianum for hosting us three years in a row, creating the perfect setting for our gathering, and to Production Pool for keeping everything running smoothly and at such a high standard. A special shoutout to our speakers for generously sharing their knowledge and experiences, introducing us to new worlds and best practices that broaden our horizons. Thank you to our attendees for the positive energy, engaging conversations, thoughtful feedback, and all the little moments that made this event unforgettable. And of course, thank to our sponsors for making it all possible, and to our volunteers for their hard work and for bringing their energy and enthusiasm every step of the way. Thanks for an unforgettable Heapcon 2024. Can’t wait to do it all again next year! 😊

Heapcon 2024 kicked off with a bang, and day one was everything we hoped for! From the gorgeous design that had everyone talking to an impressive lineup of talks covering every corner of tech, this year’s conference is already shaping up to be unforgettable. We saw old friends reunite – former colleagues, mentors, and teammates catching up under one roof – and watched as new connections were sparked. Our incredible speakers brought fresh perspectives, fueling lively discussions that attendees couldn’t get enough of. Our sponsors kept the energy high with interactive booths, exciting games, and even prize giveaways! Huge shoutout to our speakers for their inspiring talks, our attendees for their curiosity and engagement, our sponsors for making it all possible, and our volunteers for their unbeatable energy. This is what the magic of Heapcon is all about! 

We had the opportunity to hear a talk by Jeremy W. from Google, who shared valuable insights on improving web application responsiveness and enhancing user experience. One of the most important tips he shared was about using the `scheduler.yield()` function to ensure continuous task execution without interrupting user interactions. He mentioned: "scheduler.yield means that your work is what gets prioritized, aside from user interactions, which are higher priority, but that work from other task sources, once afterwards." This approach allows critical computing tasks to be executed efficiently while keeping user interactions and page rendering as the top priorities. Jeremy showed how this is crucial for improving performance and user experience, especially in modern web applications. Thanks to Jeremy for the great talk!

The second day of Heapcon 2024 is underway, bringing a fantastic lineup of inspiring talks! :) Today, Ziyang Liu will explore tech optimization, Jeremy Wagner is set to discuss ethical design and sustainable web practices, Sergey Kurnikov will cover advancements in machine learning, and Aleksandar Sabo will share strategies for effective team leadership. Alexander Chatzizacharias & Ricardo Lippolis will dive into data-driven innovation, followed by Devlin Duldulao with insights into serverless architecture. Beyond these highlights, other experts will address various tech topics – check the agenda for all the details. Between sessions, don’t miss the chance to explore sponsor booths – many have fun games and exciting prize giveaways lined up. Be sure to visit the photobooth in the lobby with a new friend, and capture a unique memory from Heapcon 2024. Finally, your feedback is invaluable! To help us improve each year, please take a moment to fill out our online survey at the Heapspace booth, where a QR code is ready for quick access. Enjoy the second day of Heapcon! :)

"Each mind is unique, and just like us, machines must learn and adapt to our individual needs." Jelena Mladenovic, a professor at the Faculty of Mathematics in Belgrade, shared valuable insights into the rapidly evolving world of Brain-Computer Interfaces (BCIs). Her talk highlighted how these technologies hold immense potential for revolutionizing medical treatments, enhancing communication, and enabling people to interact with machines in entirely new ways. Jelena emphasized a key challenge: understanding that no two minds are the same. "Every brain works differently, and that’s why these devices need to learn and adapt in a personalized way," she explained. The true power of BCIs lies not just in their ability to read brain signals, but in their capacity to continuously learn and adapt to the individual user’s brain activity. She also raised important concerns about the ethical implications. Jelena stressed that BCI technology opens up possibilities for altering cognitive functions, but it also raises questions about consent and data ownership. "Brain data is as unique as the person who owns it," she pointed out, urging the tech community to consider the long-term consequences of these developments. As we explore the potential of BCIs, it’s crucial to balance innovation with ethical considerations to ensure that individual privacy remains intact and the misuse of data is prevented.

At Heapcon all eyes are on the networking opportunities. Beyond inspiring talks, what truly makes the conference memorable are the interactions between all participants. Enjoy the relaxed atmosphere, exchange ideas, and maybe even discover potential partners or collaborators. ☺️ #Heapcon2024 #FutureIsWrittenInCode

In his talk, Mete Atamel discusses the practical challenges and considerations when working with large language models (LLMs). He shares a hands-on approach, using Google Cloud's Vertex AI, to demonstrate how multimodal models process various inputs like images, text, and more. "LLMs offer incredible capabilities, but they come with unique challenges. Even though using them can be as simple as a single API call, getting accurate and trustworthy results is a different story. These models may hallucinate and rely on outdated data, so grounding is essential."

We had an interesting session from our keynote speaker Ivett Ördög on one of the most complex aspects of software development: handling legacy code. Tackling legacy code is often seen as a necessary evil, but as we learned, it's also a testament to a company’s success. Here are some key takeaways: 1. Code Quality vs. AI Limitations: While AI has made impressive strides, it’s not yet capable of solving all quality issues in software development. Companies must continue to focus on improving code quality directly, rather than relying solely on AI solutions. 2. Rewriting Code: Weighing the Risks: Rewriting code may seem appealing, but it's a risky endeavor. Rewrites consume time and resources and can delay new feature releases, potentially harming a company's competitive edge. A careful evaluation of the need for a rewrite is essential. 3. Balancing Design and Speed: Many successful companies begin with less-than-ideal code to speed up time-to-market. The key is not major rewrites, but incremental improvements that balance quality with business goals and minimize risks. 4. Make Change Easy, Then Make Easy Change: The principle of making small, manageable changes helps in simplifying future modifications. Incremental improvements are more sustainable than large-scale overhauls and ensure the code is easier to work with in the long run. 5. Staying Competitive: The goal should be to make continuous small enhancements that keep the product ahead of competitors. It’s better to implement steady improvements than risk losing ground with large, time-consuming projects. 6. Real-World Case Studies: Ivett shared examples of companies that handled legacy code effectively: - Prezi: Transitioned from ActionScript to JavaScript, ultimately creating a product that surpassed the original. - A company improved error-handling in microservices, reducing disruptions and enhancing reliability over time. - A tailored import solution saved significant time for a client, proving that addressing specific needs can be more efficient than broad rewrites. 7. Error Management in Microservices: In microservices, it’s important to focus on the most error-prone areas first, ensuring that critical issues are resolved incrementally without disrupting the entire system. 8. Prototypes and MVPs: Building a Minimum Viable Product (MVP) first allows companies to test the market with minimal investment, helping them gauge customer interest and make informed decisions early on. Big thanks to Ivett for sharing these practical insights with us!

𝐇𝐞𝐚𝐩𝐜𝐨𝐧 𝟐𝟎𝟐𝟒 𝐡𝐚𝐬 𝐨𝐟𝐟𝐢𝐜𝐢𝐚𝐥𝐥𝐲 𝐬𝐭𝐚𝐫𝐭𝐞𝐝 𝐚𝐭 𝐌𝐚𝐝𝐥𝐞𝐧𝐢𝐚𝐧𝐮𝐦! We’re bringing together the brightest minds in tech to explore how code is shaping our tomorrow! This year, our theme, "𝐅𝐮𝐭𝐮𝐫𝐞 𝐢𝐬 𝐰𝐫𝐢𝐭𝐭𝐞𝐧 𝐢𝐧 𝐜𝐨𝐝𝐞," brings together visionary speakers and tech enthusiasts to dive into the latest in 𝐟𝐫𝐨𝐧𝐭-𝐞𝐧𝐝 𝐞𝐧𝐠𝐢𝐧𝐞𝐞𝐫𝐢𝐧𝐠, 𝐬𝐞𝐜𝐮𝐫𝐞 𝐬𝐨𝐟𝐭𝐰𝐚𝐫𝐞 𝐝𝐞𝐯𝐞𝐥𝐨𝐩𝐦𝐞𝐧𝐭, 𝐀𝐈-𝐝𝐫𝐢𝐯𝐞𝐧 𝐭𝐨𝐨𝐥𝐬, 𝐚𝐧𝐝 𝐦𝐨𝐝𝐞𝐫𝐧 𝐰𝐞𝐛 𝐭𝐞𝐜𝐡𝐧𝐨𝐥𝐨𝐠𝐢𝐞s. Every talk explores how we’re coding the world of tomorrow with insights and breakthroughs from industry experts. In this era, where code defines everything from daily convenience to groundbreaking solutions, Heapcon sets the stage for meaningful conversations and new perspectives on how we build, secure, and scale our digital future. Join us as we explore how the future is written in code—one insightful session at a time. #Heapcon2024 #FutureIsWrittenInCode

How can advanced data pipelines drive meaningful insights from GPS data? 🌍 Dusan Zamurovic, Engineering Manager at Foursquare, will dive into this topic at Heapcon 2024 with his talk, "Maximizing Airflow: Advanced Techniques for Data Pipeline Orchestration." Learn from Dušan's experience building scalable systems and transforming raw GPS pings into valuable movement patterns. Don’t miss out on optimizing your data workflows! 🚀 📅 November 7-8, 2024 📍 Madlenianum, Belgrade #heapspace #heapcon #heapcon2024 #DataPipelines #Airflow #Foursquare #AWS