FPGA

**Und was Unternehmen daraus über Sichtbarkeit, Kontrolle und Vertrauen in der KI-Welt lernen.** «Stöck, Wyys, Stich.» In der Schweiz weiss jeder sofort, was gespielt wird. Die grossen KI-Modelle kennen unser Kulturgut jedoch nur oberflächlich. Wo das Trainingsmaterial dünn ist, erfinden sie die Regeln mit voller Überzeugung. Dasselbe Problem hat fast jede Firma mit ihrem internen Fach- und Produktwissen. Remo Prinz zeigt, wie man durch strukturierte Daten Teil des KI-Training Korpus wird, verlässlich zitiert wird und dieses Fundament über ein autorisiertes Model Context Protocol (MCP) zu einer kontrollierten, vertrauenswürdigen Live-Quelle ausbaut. Ist es ausserdem möglich, ein Modell zu trainieren, das nicht nur perfekt jasst, sondern als geduldiger Mentor jeden Zug erklären kann? Ein Abend über Tradition, strukturierte Daten – und die Frage, wie Organisationen in der KI-Welt zur Quelle werden, der die Maschinen vertrauen.

**If your design feels right, how do you prove that it works?** Creatives are taught to trust their instincts, and rightly so. But intuition alone can leave your best work undefended in the room, and it cannot tell you whether a design actually changed anything. What looks better in a critique still needs to prove that it creates value in the real world. So, the real challenge is knowing what works, not just what looks good. In this keynote, **Mario Stipetic**, product designer, engineer, and product manager based in Zürich, introduces data-driven design as a practical way to use evidence without killing creativity. With over 14 years of experience designing and building SaaS web and mobile products end to end, including as Head of Product Design at a US-based fintech, Mario works at the intersection of design, product, and engineering. Join us on **June 30** as Mario shows, through real before-and-after case studies, how simple data can help teams test ideas, iterate faster, and build products people keep coming back to. **💡 What will you learn?** * Why creativity and data work best together, and how measurement helps you defend your design decisions * How simple methods like a 1–5 survey, a funnel, or a fake button can give you real proof without a big team or budget * How to set a clear hypothesis and one metric before you design, then measure before vs after * How to read what the data is telling you and turn it into sharper copy, visuals, flows, and product decisions *Connect with Mario* [LinkedIn](http://linkedin.com/in/mariostipetic) —— **📍 Where?** [Kulturhaus Helferei](https://maps.app.goo.gl/AJuPTcLTmihXxHyn9), near the Grossmünster in Niederdorf. \> Kirchgasse 13\, 8001 Zürich (Tram-Station Helmhaus) **📅 When?** Tuesday, June 30, 6:45 PM / 18:45 Uhr **⏰ What’s the schedule?** 6:30 PM → Doors open 6:45 PM → Keynote + Q&A 7:30 PM → Networking 🔗 **Discord** Get event reminders and connect with other members on Discord: ➡️ [discord.gg/PAyRBDhKkk](https://discord.gg/PAyRBDhKkk) —— **Who’s Visual Fields?** Visual Fields is a network of designers and creatives built for people to connect, support, and share with fellow experts. We have upcoming events in Zürich, Basel, and Luzern. [Check out upcoming events](https://www.visualfields.ch/) —— **Event policies** We have zero tolerance for discrimination or harassment. Be mindful not to make assumptions about gender, sexual orientation, or identity. Always ask for consent - for physical contact and work critique - and treat everyone and the space with respect. [All policies](https://www.joyed.com/experiences/ep) **Fine print (the friendly kind)** By joining, you agree that photos or videos taken at the event may be used by Visual Fields for promotional and communication purposes. See you there! 😇

Right ... meet at Begonienstr. 9, Zürich (near Nihao restaurant). Buzzer: Heiskanen. https://www.openstreetmap.org/way/42438928#map=19/47.402524/8.546013

Build //localhost:Zürich **@Microsoft** *Join us in person at Microsoft offices (The Circle)!!* **RSVP Here: https://globalai.community/e/f06j3reb** **(meetup does not host any registration so register on the above link)** For the uninitiated — Build //localhost is a globally coordinated developer event series where builders, engineers, and AI enthusiasts gather locally to demo, discuss, and discover what's being built in their own backyard. Think: real talks, real demos, real conversations — no fluff. 𝗔𝘁 𝗕𝘂𝗶𝗹𝗱 //𝗹𝗼𝗰𝗮𝗹𝗵𝗼𝘀𝘁:Zurich, 𝗲𝘅𝗽𝗲𝗰𝘁: ⚙️ Live technical sessions & hands-on demos 🤖 AI-first conversations with practitioners 🔗 Meaningful networking with Zurich's builder community 💡 Spontaneous ideas, collaborations, and maybe a few "𝘄𝗵𝘆 𝗱𝗶𝗱𝗻'𝘁 𝗜 𝘁𝗵𝗶𝗻𝗸 𝗼𝗳 𝘁𝗵𝗮𝘁" moments **Agenda:** * **14:00 - 14:15** Welcome session * **14:15 - 14:45** Keynote - Latest and greatest about GitHub Copilot and AI Agents in the SDLC - Dominique Broeglin (Microsoft) * **15:00 - 15:30** Microsoft Agent Framework at Build: From Chatbots to Agentic Workflows - Jose Luis Latorre (Swiss Life) * **15:30 - 16:00** Talk 2 - Roundtable - Dr. Alexander Wachtel (ESC Deutschland) * **16:00 - 16:30** Break * **16:30 - 17:00** Microsoft IQ and the New Enterprise Context Layer - Dr. Alexander Wachtel (ESC Deutschland) * **17:00 - 17:30** Agents at scale - Bernhard Merkle (Blockbrain) * **17:30 - 18:00** Break * **18:00 - 18:30** From Aspire to Azure Container Apps: Building Secure Containers with SBOM and CVE Checks - Damien Bowden (isolutions) * **18:30 - 19:00** Endnote - Bring Copilot to your app: real integrations with Copilot SDK - Emanuele Bartolesi (Xebia) * **19:00 - 19:30** Closing & Networking (also a small apero) The event is held in English.

**Trust, but Sandbox** Agents are becoming more autonomous, are expected to work more independently, and are being granted greater privileges. They install tools as and when they need them. And they can access all files on the system, use the browser with existing cookies, and much more. How can we allow the agent to work while still maintaining control? The talk will present three methods for isolating coding agents and compare their advantages and disadvantages. First, I’ll outline the capabilities of a local agent and the risks that come with them. In a live demo, I’ll show just how simple sandboxing can be and how great it feels to switch to ‘YOLO mode’ without a second thought and let the agent work on his tasks. Finally, we’ll take a look at some practical features that a separate environment for agents can offer. **Michael Krämer** is working in software development for over 20 year and works as a Software Architect at INNOQ. He is very commited to work out component designs with clearly defined responsibilities and to find appropriate solutions for requirements. Besides that he is engaged in Machine Learning, the integration of ML models in production environments and works as a trainer for Software Architecture.

**In diesem Beitrag schildert der Collaboard-Geschäftsführer Michael Görög die Aufholjagd seines kleinen Teams gegen etablierte internationale Wettbewerber, die teils Jahre Vorsprung hatten. Und wie daraus eine starke, souveräne und sichere Alternative zu Mural und Miro wird, die beim Gebrauch auch noch Spass macht.** Collaboard begann mit dem Aufstieg agiler Arbeitsformen und aus dem Wunsch, interaktiv an elektronische Whiteboards zu arbeiten. Der ursprüngliche Fokus lag daher auf einer Whiteboard-Software für interaktive Displays, um Meetings und Workshops im physischen Raum digitaler und interaktiver zu gestalten. Aber mit der Pandemie veränderte der Markt sich radikal. Über Nacht brach der ursprüngliche Product-Market-Fit von Collaboard in sich zusammen: Statt digital unterstützter Zusammenarbeit im gemeinsamen Büro oder Sitzungszimmer war jetzt Remote-Work angesagt. Das Collaboard-Team stand damit vor der radikalen Herausforderung, innert kürzester Zeit aus der elektronischen Whiteboard-Software ein digitales Whiteboard für Remote-Zusammenarbeit zu entwickeln. ••• [Michael Görög](https://www.linkedin.com/in/michael-gorog/) ist Co-Founder von [Collaboard](https://www.collaboard.app), einer in der Schweiz entwickelten Online-Whiteboard-Lösung für visuelle Zusammenarbeit. Mit über 16 Jahren Erfahrung in der Softwarebranche treibt er Innovationen im Bereich visueller Teamarbeit voran. Sein besonderer Fokus liegt dabei auf Sicherheit, Datenschutz und dem Einsatz moderner Technologien wie KI.

Coders Monthly is a regular gathering of programmers in Zurich interested in honing their craft. It takes place in Zuerich at Cafe Bar Pluesch (https://cafe-pluesch.ch) on the first Thursday of every month. For the members list, some pics and more, see the Coders Only (https://codersonly.org/events/coders-monthly) site. Hope to see you there! PS: Please join our Discord https://discord.gg/jWUZsKQvrz for any last minute announcements.

Hares (someone to set the route) are NewBalls, SNOBalls, and All Hips Matter. Come join us! **Where**: See location. Typically within Zurich Zone 10. Look for English speakers wearing running gear. **When:** We'll meet at 6:50 pm to start at 7:00 pm. There's a charge of CHF 5 for participating, which covers all you can drink beer/cider (and non-alcoholic alternatives) plus snacks afterwards. **All Welcome:** If you are new, curious, and want to get out of the house, stretch your legs, talk shit, enjoy a drink, and hopefully a laugh, then here's what you need to know: • We are a social running/walking group, physical fitness is not required, but a sense of humour is useful. • The aim is to find the '*Beer Stop*' at the end of the '*trail*' and have fun. • 'True Trail' is typically 4-7 km. Some will go less (especially those walking or walk/running), and some over-achieving fast runners will go further. • The '*Hare*', which changes each event (along with location), cleverly lays the trail (typically with flour marks) so that everyone reaches the end at the same time. • There is always a place to deposit your bag and sometimes a place to change. • **After Trail**, we group for \~20-40 minutes for drinks, snacks, and some fun. This '*Circle*' is typically outside, so leave something warm in your bag in winter (and bring a flashlight when it's dark). If you don't want to stand around drinking afterwards, you can head home. We occasionally all go to a restaurant afterwards. **Why bother leaving your warm & dry sofa?** If you enjoy getting outside, being around people, and doing something a bit different, you will probably enjoy hashing. Exercise is important, and this is probably the least boring way to get it. 'Hashing' and the idea of some physical activity to work up a thirst, have been around for almost 90 years and there are groups (*Hashes*) in most major cities. Its humor and 'traditions' can be explored on the internet - **or you can just come along and find out.** *See you there.* We are of many nationalities, all ages, full of misfits and idiots (at least there is no one taking themselves seriously!) and always welcome newbies 'Virgin Hashers' and visitors from other groups. For more about Zürich Hash House Harriers, check out [www.zh3.ch](http://www.zh3.ch/) ( [http://www.zh3.ch/](http://www.zh3.ch/) )

Two complementary takes on what actually decides whether AI delivers value in an organization. One on why role-based structures are the real enabler, the other on what happens to teams when AI starts resolving the tickets. Here's what's planned for the evening: 17:30 - 18:00 - Arrival and first drinks 18:00 - 18:45 - AI Value Creation Starts Beyond Technology by Ralf Günthner 18:45 - 19:30 - When the Bot Takes the Ticket: Keeping Teams Effective as AI Rewrites the Rules by Dagmar Muth 19:30 - 20:30 - Networking, Open Space & More drinks\* **AI Value Creation Starts Beyond Technology** Many companies are currently investing heavily in AI agents, copilots, and automation, yet the expected gains in productivity and innovation often fail to materialize. The reason is rarely technological, but structural: classic hierarchical organizational models are not designed to collaborate effectively with learning, adaptive systems. This keynote explains why role-based work is the decisive enabler for successful AI adoption. In role-based organizations, responsibilities are clearly defined, decision rights are decentralized, and collaboration is dynamically organized, exactly the conditions under which AI agents can deliver maximum value. Drawing on current research (including McKinsey, MIT CISR, and Harvard Business School) and practical examples from industry, mechanical engineering, and chemicals, the keynote shows how companies can combine AI agents with role-based structures to: * drastically shorten implementation timelines * increase acceptance and trust in AI * enable continuous optimization instead of one-off pilot projects * significantly improve the ROI of AI investments The core message: AI does not transform organizations, organizations determine whether AI creates impact. Those who invest today in role-based ways of working, shared leadership, and decentralized decision logic build the operational foundation to not only introduce AI, but scale it strategically. **Ralf Günthner** Ralf Günthner is a transformation expert, entrepreneur, and lecturer with over 25 years of experience in digital transformation, AI, IoT, and organizational development. In his keynotes, he shows why AI value creation does not emerge from algorithms, platforms, or tools, but from leadership, culture, decision-making logic, and effective collaboration. He combines deep technology expertise with systems thinking, neuroscience, and hands-on transformation work. As Co-CEO of Team-Factory GmbH, he supports companies in anchoring AI strategically, creating orientation, and delivering sustainable impact, beyond hype and technology fixation. **When the Bot Takes the Ticket: Keeping Teams Effective as AI Rewrites the Rules** AI agents are starting to resolve incidents, write code, and close tickets. That's great for throughput, but what happens to the humans in the team? Motivation, ownership, skill development, and psychological safety don't disappear as team development topics just because productivity metrics go up. In this talk, Dagmar Muth draws on her work with engineering and product teams navigating AI-driven transformation. She shares what it takes to make AI integration work for the whole team: how to preserve ownership and purpose, keep accountability clear, and build the patterns that distinguish teams that thrive from those that merely adapt. You'll leave with a clear mental model for thinking about team health in AI-augmented setups and concrete questions to bring back to your own team or organization. **Dagmar Muth** Dagmar Muth is a systemic organizational developer and agile coach with 20 years of experience guiding teams and companies through transformation. As a co-worker of pragmatic solutions ag in Zürich, she supports organizations in finding the right setup, processes, and culture to sustain performance from vision to go-live, increasingly in contexts shaped by AI. She works with cross-functional tech and product teams, coaches leaders, and lectures on agile project management at SIB. Her particular focus right now: how organizations integrate AI without losing what makes their teams actually work. Many thanks to our **sponsors**: * Catering Sponsor: **DevOpsDays Zürich** https://www.devopsdays.ch/ * Location Sponsor: **Digicomp** https://www.digicomp.ch/ \* contact us min. 3 days prior to the event if you prefer a vegan option.

Bienvenida al del club de lectura en Zürich! Un espacio para conectar, compartir y conocernos entre lectoras que viven en Zürich y alrededores.
 El libro elegido para la próxima reunión es: **Un mundo Feliz de Aldous Huxley** así que tienes hasta el **04 de Julio** para leerlo y unirte a nuestra discusión. El enfoque lector del club es amplio: ficción, fantasía y novelas de todo tipo **¿Para quién es?** Para personas mayores de edad que quieran participar en un club de lectura. Será un espacio relajado, basado en el respeto y el amor por los libros. **¿Qué necesitas?** - Una libreta, lápiz o bolígrafo - Una manta - Agua y snacks para ti y/o para compartir. - El libro en físico o digital (opcional) Antes de hablar del libro haremos algunas actividades para romper el hielo, compartiremos tips para conseguir un hábito lector y hablaremos de recomendaciones lectoras. Como **recomendación** puedes tomar notas del libro, si eso te ayuda a recordar mejor. El debate no será académico, será sobre las opiniones personales de personajes, narrativa, historia, etc. Confirmaré unos días antes si, por motivos de clima, cambiamos a una cafetería. Si no llueve, nos vemos en el lugar indicado. (Revisa los mensajes del evento un par de días antes) Por favor sé puntual😊

Want to be a speaker? submit your talk to our Call for Presenters!!! https://sessionize.com/azure-cbus-2026/

You can sleep in on the Sesquicentennial (observed), we’re shifting to the following Friday. But join the local UX community for a chat and some biscuits to catch up. If you were in CA for Figma Config, tell us all the lines and you can tell me how I can animate an SVG without coding it by hand, and why I’d even want to. Thanks to the team at Nationwide of supporting the group.

Bring your Raspberry Pi, Arduino, microcontroller, or any other electronic project and join fellow electronics makers for a night of creativity and collaboration! This session is open forum to share your current projects—whether complete or in progress, it’s all interesting! Whether you’re deep into embedded systems, exploring new ideas, or just getting started, you’ll find a welcoming space to collaborate, share, and get inspired. **New to electronics or curious about tinkering?** You’re absolutely welcome. If you’re a beginner and want to experiment, I’ll have a couple of starter kits available so you can try things out—whether that’s blinking your first LED, putting something on a display, or experimenting with simple sensors. No pressure and no experience required—just an interest in learning and building. While we continue to pursue a more permanent venue for this Meetup, we’ll be using public library facilities based on availability.

Most people use LLMs by feel: ask a question, read the answer, decide whether it “seems good,” and move on. That works for casual use. It does not work when you are building software, automating workflows, writing important documents, or relying on AI for anything that needs to be repeatable. In this talk, we’ll look at how to improve and evaluate the inputs and outputs of LLMs using practical measurement techniques. We’ll cover how prompt changes affect results, how to compare outputs, how to build simple evaluation sets, and how math-based methods like similarity scoring can help you move beyond guesswork. This will be beginner-friendly, so even if you don't know anything about AI, you should get something out of it. However, this will be a little more technical than our intro talks. You do not need to be an AI researcher, but programmers and technically curious attendees will get a lot out of it. We’ll cover: * Why “it looks good” is not enough * How to improve prompts by changing the input, context, and constraints * How to compare LLM outputs more systematically * Basic evaluation techniques for accuracy, consistency, and usefulness * How embeddings, cosine similarity, and scoring can help evaluate results * Where automated evaluation works — and where humans still need to stay in the loop By the end, you’ll have a practical mental model for treating LLMs less like magic and more like systems you can test, measure, and improve. LOGISTICS AND PARKING: The talk starts at 7:00 PM. The first half hour is reserved for everyone to get set up and mingle. Free pizza and drinks! The cheapest parking option is to find street parking, which will only cost you a few bucks. Otherwise, park in the nearby veteran's museum lot for $8. It's highly recommended you avoid the nearby $15 garage parking.

We have moved to guild.host. Look us up there. https://guild.host/events/buckeye-cocoaheads-rcavwd

Come join us for lunch together at Bendi Wok, one of my favorites!