Data Visualization

Join us for a 2-day hybrid hackathon in Tirana where developers, students, and people interested in AI come together to learn, build, and show their AI projects. Get practical experience, support from Google Developer Experts (GDEs), and take part in a fun demo session with prizes and giveaways. * **What participants will experience:** * Masterclasses from speakers * Team collaboration and project building * Mentorship from industry experts and GDEs * Real-world project development from idea to demo * Networking with the local tech community * Fun challenges, prizes, and giveaways **Day 1 – Online (20 May from 5pm): Start Building** * Join the welcome and introduction session * Learn about AI through masterclasses * Form teams and share project ideas * Start building their AI solutions **Day 2 – Onsite (22 May at Piramida from 10am): Build, Improve, and Present** * Panel discussion with industry experts * Learn about AI, cloud technologies, and real-world use cases * Continue building your projects with mentor guidance * Prepare and present your solutions to a panel of judges 👉 Fill out this form with your project idea and/or team details: [https://forms.gle/8cui8JKNWEp82tnaA](https://forms.gle/8cui8JKNWEp82tnaA) If you don’t have a team yet, don’t worry — you can connect with others and find team members through our Discord community: [https://discord.com/invite/BgCdCmCuGv](https://discord.com/invite/BgCdCmCuGv) **Grand Finale: Demo & Celebration** * The event ends with: * Team project presentations * Feedback from judges and mentors * Prize announcements and giveaways * A closing celebration of participants’ achievements **A chance to learn, build, and grow in AI**

Со големо задоволство ве покануваме на новиот PyData собир, кој ќе се одржи во среда, на **20 мај 2026 година**, со почеток во 18:00 часот во Base42. Овој пат ќе имаме можност да проследиме предавање на тема **„On an Inefficiency in the Linear Attention Projections, and How to Fix It“**, кое ќе го одржи **м-р Марко Карбевски**. Предавањето ќе понуди интересен спој на длабоки теориски увиди во Transformer архитектурите и лично искуство од независно ML истражување. Во рамки на презентацијата ќе биде претставено истражување кое покажува дека Query тежините во standard self-attention механизмот можат целосно да се елиминираат без нарушување на функционалноста на моделот. Со ова се намалуваат параметрите и пресметковната комплексност за повеќе од 8%, а наместо редундантен линеарен слој се користи мал и ефикасен нелинеарен модул кој овозможува постабилен тренинг и подобри резултати кај GPT-стил модели. Дополнително, ќе стане збор и за математичките основи зад истражувањето, вклучително и нови сознанија за улогата на skip connections кај MLP мрежите и нивното влијание врз функциите што моделите можат да ги репрезентираат. Покрај техничкиот дел, Марко ќе зборува и за предизвиците на независното ML истражување — од експерименти и hyperparameter пребарувања, до искуството од учество на DeLTa и GRaM работилниците на овогодинешниот ICLR во Рио де Жанеиро 🇧🇷. ICLR е широко признаена како една од трите најпрестижни научни конференции во светот во областа на машинското учење. **Марко Карбевски** е магистер по теориска математика со осумгодишно искуство како Data Scientist. Дипломирал во областа на теориската математика, а магистерските студии ги завршил на Сорбона во Франција. Неговите интереси се насочени кон теориските аспекти на машинското учење и фундаменталните ML архитектури.

Join us for a 2-day hybrid hackathon in Tirana where developers, students, and people interested in AI come together to learn, build, and show their AI projects. Get practical experience, support from Google Developer Experts (GDEs), and take part in a fun demo session with prizes and giveaways. * **What participants will experience:** * Masterclasses from speakers * Team collaboration and project building * Mentorship from industry experts and GDEs * Real-world project development from idea to demo * Networking with the local tech community * Fun challenges, prizes, and giveaways **Day 1 – Online (20 May from 5pm): Start Building** * Join the welcome and introduction session * Learn about AI through masterclasses * Form teams and share project ideas * Start building their AI solutions **Day 2 – Onsite (22 May at Piramida from 10am): Build, Improve, and Present** * Panel discussion with industry experts * Learn about AI, cloud technologies, and real-world use cases * Continue building your projects with mentor guidance * Prepare and present your solutions to a panel of judges 👉 Fill out this form with your project idea and/or team details: [https://forms.gle/8cui8JKNWEp82tnaA](https://forms.gle/8cui8JKNWEp82tnaA) If you don’t have a team yet, don’t worry — you can connect with others and find team members through our Discord community: [https://discord.com/invite/BgCdCmCuGv](https://discord.com/invite/BgCdCmCuGv) **Grand Finale: Demo & Celebration** * The event ends with: * Team project presentations * Feedback from judges and mentors * Prize announcements and giveaways * A closing celebration of participants’ achievements **A chance to learn, build, and grow in AI**

Со големо задоволство ве покануваме на новиот PyData собир, кој ќе се одржи во среда, на **20 мај 2026 година**, со почеток во 18:00 часот во Base42. Овој пат ќе имаме можност да проследиме предавање на тема **„On an Inefficiency in the Linear Attention Projections, and How to Fix It“**, кое ќе го одржи **м-р Марко Карбевски**. Предавањето ќе понуди интересен спој на длабоки теориски увиди во Transformer архитектурите и лично искуство од независно ML истражување. Во рамки на презентацијата ќе биде претставено истражување кое покажува дека Query тежините во standard self-attention механизмот можат целосно да се елиминираат без нарушување на функционалноста на моделот. Со ова се намалуваат параметрите и пресметковната комплексност за повеќе од 8%, а наместо редундантен линеарен слој се користи мал и ефикасен нелинеарен модул кој овозможува постабилен тренинг и подобри резултати кај GPT-стил модели. Дополнително, ќе стане збор и за математичките основи зад истражувањето, вклучително и нови сознанија за улогата на skip connections кај MLP мрежите и нивното влијание врз функциите што моделите можат да ги репрезентираат. Покрај техничкиот дел, Марко ќе зборува и за предизвиците на независното ML истражување — од експерименти и hyperparameter пребарувања, до искуството од учество на DeLTa и GRaM работилниците на овогодинешниот ICLR во Рио де Жанеиро 🇧🇷. ICLR е широко признаена како една од трите најпрестижни научни конференции во светот во областа на машинското учење. **Марко Карбевски** е магистер по теориска математика со осумгодишно искуство како Data Scientist. Дипломирал во областа на теориската математика, а магистерските студии ги завршил на Сорбона во Франција. Неговите интереси се насочени кон теориските аспекти на машинското учење и фундаменталните ML архитектури.

**Important time note:** Please plan on arriving between 5:30 and 6:00 as the elevators lock after 6 and you'll need to message us and we'll need to come get you. The building address is 4450 Bridge Park The entrance is 6620 Mooney St, Suite 400 You will need to scan your ID at the door to get a visitor badge. **Abstract** TBD **YouTube Link** TBD

The June Ohio North Database Training user group meeting will be held on **June 2nd, 2026 at 5:00PM**. This will be a **HYBRID** event and we will be joined online by **Sam Nasr.** You're welcome to come meet in-person at our meeting location, the offices of Improving at **[6000 Freedom Square Dr,](https://www.google.com/maps/place/Improving/@41.4004167,-81.6614462,17z/data=!3m2!4b1!5s0x8830e5b8255c5919:0xd8297060eb68fe04!4m6!3m5!1s0x8830dc7a0fe35dc9:0xbfc4710ecadfc5c!8m2!3d41.4004127!4d-81.6588713!16s%2Fg%2F1hm3hkqp3?entry=ttu&g_ep=EgoyMDI1MDQzMC4xIKXMDSoASAFQAw%3D%3D)** **[Unit 110,](https://www.google.com/maps/place/Improving/@41.4004167,-81.6614462,17z/data=!3m2!4b1!5s0x8830e5b8255c5919:0xd8297060eb68fe04!4m6!3m5!1s0x8830dc7a0fe35dc9:0xbfc4710ecadfc5c!8m2!3d41.4004127!4d-81.6588713!16s%2Fg%2F1hm3hkqp3?entry=ttu&g_ep=EgoyMDI1MDQzMC4xIKXMDSoASAFQAw%3D%3D)** **[Independence, OH 44131](https://www.google.com/maps/place/Improving/@41.4004167,-81.6614462,17z/data=!3m2!4b1!5s0x8830e5b8255c5919:0xd8297060eb68fe04!4m6!3m5!1s0x8830dc7a0fe35dc9:0xbfc4710ecadfc5c!8m2!3d41.4004127!4d-81.6588713!16s%2Fg%2F1hm3hkqp3?entry=ttu&g_ep=EgoyMDI1MDQzMC4xIKXMDSoASAFQAw%3D%3D)** [Teams Link ](https://teams.microsoft.com/meet/215625630917731?p=vxNb3zudhNj0S6Cg0v)if anyone needs it after RSVP-ing for in person. If you would like to subscribe to our email list outside of Meetup, we have changed platforms recently and you will need to register [here in Kit ](https://ohio-north-data-training.kit.com/b8f036f615)instead to receive emails. Agenda: **5:00 PM EST**: Online and in-person meeting begins with a social hour. This is an unstructured hour where you can join us to catch up and meet other group members before the session starts. There will be food brought in for in-person attendees. **6:00 PM EST**: Elections, announcements, followed by our feature presentation. See below for presentation details. **7:30 PM EST**: Optionally after the main presentations, the in-person crowd may go out for snacks and drinks at a local establishment. We hope to see you there! Session Abstract ### SQL Server 2025: The Developer’s Revolution Explore SQL Server 2025’s new dev features -JSON type, Regular Expressions, and Change Event Streaming for real-time pipelines. See how REST API and Data API Builder turn your data into secure, high-performance APIs. Packed with demos, this session is perfect for developers, DBAs, and architects \*Please note, that we will be using Microsoft Teams for the online portion of this meeting. You may want to join a few minutes early to ensure you do not have any issues. If you are attending in person, there are large TVs at the office, and you do not need to bring a laptop or use Teams.

NFT's are here to stay folks! This is a group for like minded people interested in understanding, leveraging, using, creating for, profiting from, trading too i suppose, NFT's.. everything around them, complexity, fear and exploits, best practices and more. **PLUS** This group will talk AI ART tools, techniques, artists, video, audio, prototypes and more in the AI assisted production space- ART specifically, but we can get into any aspect of some of the cooler things happening in AI in general.

**Abstract:** This is a practical playbook for building a production-grade data lakehouse. It walks through foundational principles — naming conventions, least-privilege access, automated CI/CD testing — before diving into medallion architecture. Furthermore, metadata-driven design patterns show how configuration tables and dynamic notebook orchestration eliminates hard-coded pipelines. The deck covers star schema modeling, guidance on choosing between Spark, Pandas, and SQL, and data quality enforcement using DQX with YAML data contracts. Finally, we dive into security best practices and performance optimizations. **Host:** Justin Shea, Mehdi Jeddi, Erik Pak, and Sou-Cheng Choi **Talk Format:** This is a hybrid event. To attend online, join us on Zoom here at 6pm: https://iit-edu.zoom.us/j/89379230295?pwd=NdETyE5sdYuSrvsrBZXSBFkUESBVkg.1 Meeting ID: 893 7923 0295 Passcode: 5t5WYn **Sponsor:** Adyen, UIC College of Business, and PyData Chicago co-host this event. UIC will provide the meeting site. Adyen will sponsor pizza and soft drinks for the onsite participants. **Address:** University of Illinois - Chicago, Douglass Hall, Room 220, 705 S Morgan St, Chicago, IL 60607 **Logistics:** “UIC Douglass Hall” is recognized on Google Maps, which can guide you through campus. Once you arrive, proceed to the second floor, room number 220

Welcome to Casual Boardgames, where we enjoy classic tabletop games and social deduction games while bonding over good food, drinks, and great conversations. 
We started this group to meet new people and make new friends, and bonding over games in a relaxed atmosphere is a great way to do that. We currently meet near route 23 and Polaris Parkway, and this is close to areas like Powell, Lewis Center, Worthington, and parts of Westerville and Columbus. 
Feel free to bring your own games or play one of the many games our members bring. If you are inexperienced, we will help you learn. 
Here are just a few examples of the kind of games we play. 
Tabletop/board games: Splendor, Catan, Azul, Dominion, 7 Wonders… 
Social deduction games: Code Names, Chameleon, Werewords… 
IMPORTANT: 
1. While we love playing a variety of games and competing, we are not just about the games. We interact and talk while playing, and this leads to a lot of laughter and fun. If you just want to compete and focus solely on the game, then this is probably not the group you are looking for. 
2. Many people join groups like this and never (or rarely) show up after weeks or months. If you join and never really come, we will eventually remove you from the group as a courtesy to our members. Why? Because limiting group conversations (on the app) to regular members makes communication and planning much easier. Also, while it may be rare, it protects members from people who join because they are interested in following a person instead of having a real interest in our group. If you get removed, it is just because you haven’t come, and we follow this protocol as a courtesy to our regular game players.

Let’s meet and wander the galleries! General admission on Sundays is free.