Deep Reinforcement Learning

Hello fellow Symfonians, we are thrilled to announce the first user group meeting of the year, featuring two insightful talks that you **won't want to miss**! This time we are hosted by c-base (Rungestrasse 20 10179 Berlin). Come by and enjoy an evening of learning, networking, and socializing with fellow Symfony and PHP developers. **Agenda:** 18:30: Doors open 19:00: Welcome and Introduction 19:20: **Talk 1: "Agentic coding with OpenCode" by Damijan Cavar** 19:50: Break & Snack 20:00: **Talk 2: "Composer Deep Dive" by Nils Adermann** 20:40: Socializing **Talk Details:** **Talk 1: "Agentic coding with Opencode" by Damijan Cavar** Summary (ChatGPT generated): AI coding assistants are evolving beyond “vibe coding” toward **agentic, spec-driven development**. This talk explores why prompt-based workflows break down at scale and how structured specifications create a reliable foundation for autonomous coding agents. You’ll see how **OpenCode**, an open-source, model-agnostic coding agent, uses planners, implementers, and reviewers to turn specs into verifiable code while keeping context small and repeatable. The session also covers practical security considerations, sandboxing strategies, and real-world lessons from using agents on non-trivial projects. **Talk 2: "Composer Deep Dive" by Nils Adermann** This talk will take the audience on a deep dive into Composer. We'll start on the surface with the big picture of Composer (how exactly do require/update/install relate to json/lock/vendor), followed by commands and arguments not widely known and used (e.g. bump, why/why-not, update options like --with-all-dependencies, --minimal-changes, and --ignore-platform-req=php+ and what exactly the plus means). We'll then advance to the underlying concepts like the platform repository, security concepts around Composer plugins and answer questions like "Why is there a hash in the lock file and how to resolve conflicts?", or "Why do repositories need to be defined on the root composer.‎ json?" and finally reach into dependency resolution with a quick explainer on how Composer uses a SAT solver and boolean algebra to decide which packages to install. Back above water, the talk will end with an outlook on changes we're planning for Composer and [packagist.org](http://packagist.org) in the future. Don't miss these insightful talks, engaging discussions, and networking opportunities. We can't wait to see you at the **February Symfony User Group**!" If you have any questions or accessibility requirements, please reach out to us. Also, if your company wants to be the next host for the User Group, just let us know!

Get hands-on with Google Agent Development Kit (ADK), Google’s framework for building, evaluating, and deploying AI agents. You'll progressively learn to design advanced agents and train them to play the text-based adventure game “The Garden of the Forgotten Prompt”. Get ready to explore a mysterious world, solve puzzles, and uncover secrets! Agenda: ▶️ 6:00 pm: Doors open with networking and snacks ▶️ 7:00 pm: Opening Keynote ▶️ 7:15 pm: Workshop (Laptops are required. Please charge them in advance!) Abdelfettah Sghiouar (Abdel) - Cloud Developer Advocate @ Google Cloud Daniela Petruzalek - Senior Developer Relations Engineer @ Google Cloud ▶️ 9:00 pm: Network & Drinks ☕ ▶️ 9:30pm: Doors close Google Cloud Credits We’re planning to offer Google Cloud credits to attendees so you can explore and experiment on GCP during the event. Activation details and instructions will be shared on-site. Agenda --- Speakers Daniela Petruzalek - Google (Senior Developer Relations Engineer @ Google Cloud) Abdel Sghiouar Hosted By Yevgen Batovskyi, Organizer Esra Kadah, Organizer Senior App Developer | Author of "Flutter Design Patterns & Best Practices" | Co-Organizer of Flutteristas, Flutter Berlin, GDG Cloud Berlin, Flutter Community AI Circle | ‍ Antigua Mobile Hany Elfouly, Co-founder & CEO Mohamed Islem Ayari, Organizer --- Partner Google (https://blog.google/intl/de-de) --- Complete your event RSVP here: https://gdg.community.dev/events/details/google-gdg-cloud-berlin-presents-hands-on-with-agent-development-kit-adk-google-office/.

Let's talk Android! This month, we are at SumUp and are welcoming these amazing opportunities: **Android AI Workflows** - *Frank Tamre* I will share a practical, what i sue set of AI workflows I use in Android projects to ship faster without sacrificing code quality. This isn’t a “prompt engineering” talk; It’s a set of repeatable patterns that can integrate into Android engineering. **Crafting Idiomatic APIs with Kotlin and Compose** - *Ash Davies* In this talk, we’ll dive into the principles of idiomatic Kotlin and explore how they apply to designing Compose APIs. You’ll learn how to leverage Kotlin’s language features, from DSLs and inline functions to advanced type safety and composable conventions, to craft APIs that are a joy to use. By mastering these techniques, you can deliver APIs that not only “work” but truly feel like Kotlin. **Your Talk here** - *You* *{Insert your talk topic here}* Afterwards, we welcome you to stay and socialize (or sit in the corner listening) with or against our community. Happy Coding / Prompting / Designing / Leading / Working.

Welcome to an intimate and powerful breathwork experience inspired by Holotropic principles. We create a safe, grounded and transformative environment for emotional release, clarity and deep inner connection. This circle is limited to **4 participants** to ensure personal guidance, emotional safety and comfort. *** ## **Session Structure** **1\. Welcome & Introduction \(30 min\)** * What Holotropic-inspired breathwork is * How the session works * Safety guidelines and recommendations for a free, supportive experience **2\. Active Breathwork Journey \(90 min\)** A deep breathing experience guided by a carefully curated musical sequence designed to support inner exploration. During this phase, participants may experience: * Body sensations, shaking, warmth, spontaneous movements * Emotional release (crying, laughter, sighs) * Images, memories, visions or states of expanded awareness **3\. Integration Circle \(60 min\)** A voluntary space to share and express your experience Suggestions for emotional and somatic integration Compassionate support, presence and grounding **Total duration: \~3 hours** *** ## **What to Bring** Please bring the following for comfort and immersion: * **Comfortable clothing** * **Eye mask** * **Water bottle** * **Blanket** * **Pillow / cushion** * **Yoga mat** (We have a few extra items if needed.) *** ## **Contraindications** Not recommended for people with: Cardiovascular conditions, epilepsy, pregnancy, recent major surgeries, or severe psychiatric diagnoses. If you’re unsure, feel free to message us. *** ## **Who We Are** We are **Paula & Javi**, certified breathwork facilitators and founders of **Holostar33**. To learn more about us and hear testimonials from other customers, follow us on Instagram: @holostar33 We are new in Berlin and building a conscious community centered around breathwork, embodiment and healing. We look forward to breathing with you. **Holostar33 – Transformational Breathwork & Community**

Mit dem Titel "Philosophie 2.0" wird nicht zu viel versprochen. Nicht alles was mit 'Philosophie' bezeichnet wird, hat damit auch zu tun. Das gilt allgemein. Wörter werden zu oft nicht präzise gebraucht, aber wenn man alles mit beliebigen Wörtern bezeichnen würde, könnte man nichts mehr genau verstehen, könnte nichts mehr genau begreifen. Also blieben die Irrtümer bestehen. Das hätte mit Philosophie aber definitiv nichts mehr zu tun. Was also ist das Gegenteil belanglos-vieldeutigen Wörterverdehens mit hochassoziativen Sprüngen bei denen allerlei übersehen wird und ergo das Wesentliche nicht mehr als solches klar erkannt werden kann? Maximal exakte Begriffsklärung. Und das geht! Das geht über einen systematischen Aufbau aller Kategorien - gestützt über einen Algorithmus, welcher Kategorienfehler automatisch vermeidet und mithilft, das Ganze im Blick zu behalten. Dass dies für uns Personen gar nicht so einfach ist, sieht man daran, dass jede Person (auch Du) irgendwann irgendwelchen Irrtümern aufgesessen ist. Jede Person, die sich also für irgendetwas ernsthaft (nicht wegen blanglosem Amüsement, oder wegen ideologischer Rechthaberei) interessiert und ergo dafür offen ist seine Auffassungen weiterzuentwickeln, ist bei dieser Veranstaltung goldrichtig. Ablauf: Bei unserer Diskussion geht es dabei nicht um irgendein festes Thema, sondern jeder schreibt ein Wort in seinem Interessengebiet auf einen Zettel und wir ziehen diese paarweise für je eine Runde. Die beiden Wörter haben gemeinsame Kategorien, aber es gibt auch Unterschiede. Wer diese nicht alle kennt, kann nicht behaupten, dass er tatsächlich begriffen hat, wovon er meint, dass es ihn interessiere; dass er definitiv wüsste wovon er spricht. Dieses Problem hatte schon Sokrates erkannt. Nur haben wir jetzt einen Algorithmus, welcher zu absoluter Genauigkeit verhilft um diese Irrtümer loszuwerden. Daher 'Philosophie 2.0'. Siehe dem Bild oben werfen wir die Bedeutung (die Menge aller Kategorien einer Begrifflichkeit) per Beamer an die Wand. Wir sind nicht in der Schule, d.h. es gibt keinen Zeitdruck, keine Noten, keinen Wettbewerb und keine Profilierungsmotive, sondern jeder hat alle Zeit der Welt um über die schematisch dargestellte Bedeutung nachzudenken - und diese zu hinterfragen. Die grafisch dargestellten Bedeutungen des Algorithmus sind nicht perfekt. Es geht nicht darum irgendetwas zu akzeptieren, sondern andersherum: die Bedeutung anzugreifen und zu verbessern. Jeder kann sie ändern! Die Diskussion ist also eröffnet. Jeder muss Gründe angeben. "Kann ein Walfisch ein Wellensittich sein?" \~> Natürlich nicht, denn Säugetiere sind keine Vögel. "Kann ein Fenster ein Getränk sein?" \~> Jeder weis, das ist unmöglich, denn Festkörper sind keine Flüssigkeiten. "Kann eine Demokratie eine Republik sein?" "Kann eine Vermutung eine Meinung sein?" "Ist etwas zu begreifen das Selbe wie etwas zu verstehen?" Nichts regt dermaßen zum nachdenken an wie die Bedeutung der Begrifflichkeiten algorithmisch stabilisiert zu systematisieren. Und wenn wir uns nicht einig werden? Die Diskussion läuft auch nach der Veranstaltung weiter. Jeder kann online von zuhause die Diskussion fortführen - mit dem selben mitdenkenden Algorithmus. Man könnte sagen, es sei ein neues Gesellschaftsspiel - ohne Wettkampf, sondern in kollektiver Bereicherung an Klarheit, von der alle gigantisch profitieren, obschon jeder nur einen sehr kleinen Teil beiträgt. Das "Spiel" funktioniert auch mehrsprachig, denn der platonische Raum der Begrifflichkeiten ist der eindeutige Raum, auf den das Vokabular einer Sprache nur referiert. Man kann auch von Zuhause aus online mitmachen. Den Zugang können wir dann über die Kommentare hier organisieren. Bring deinen Laptop mit, dann wird es einfacher... Dieses Spiel macht klug: Schach trainiert zwar das Konzentrationsvermögen, hat aber mit der Wirklichkeit nichts zu tun. Dieses Spiel hingegen schon, denn genau mitdenken zu können setzt voraus genau zu unterscheiden. Wer die Bedeutung der Wörter sauber ausdifferenziert wird immer klüger. Mach mit!

Hello! I’m organising a coworking + socialising session for all of us who are working on projects (e.g. our own creations, freelance or remote work) but don't feel like sitting alone at home. To ensure that we have time for both socialising AND working, we will be using the so-called “Pomodoro” technique: **50 mins deep work** (silent), followed by **10 mins break** to socialise. And then repeat! You are welcome to join whatever your personal background is, and whatever project you are working on. I (the host) speak English and German, so either language is fine. ——————————————————- AGENDA: * **Short Intro** Tell us who you are and share something about what you will be working on today * **Pomodoro Session** \- 3x \(3h\) Each Pomodoro session consists of 50 mins focused “deep work” (no talking, no interruptions, **phones on silent mode**) -> then 10 mins break: a chance to share your progress with others -> Repeat * **15 mins Wrap Up** Celebrate what you achieved and get to know others ———— The location is **Espresso House near Checkpoint Charlie** (they have multiple outlets - so be sure to go to the correct one!) Wifi + power plugs are available - but best to bring a fully charged laptop, just in case You **do not** have to pay anything. But since we are using the café space, it is common courtesy to buy at least one drink (coffee costs roughly **4 Euros**).

**RSVP on the [343 Labs Website](https://343labs.de/events/open-house-masterclass-11/) to reserve your seat!** Join us for a free DJ Masterclass and open house event. During the workshop, Speaker Honey will explain her concept of live flow; making each set unique with live edits, hot cue techniques and creative mixing. Enter a flow state and maximize your creativity with these secrets from the toolbox of a professional touring DJ. This masterclass is suitable for beginner and experienced DJs alike. During the event, 343 Labs will be opening its doors for an Open House, where attendees can tour the facilities, learn about 343 Labs courses, and meet some of the instructors. This is a great way to get a feel for the community and programs at 343 Labs, meet fellow producers, and ask any questions you may have about our school.

**Week 6 Focus: Publishing Your Work** Each 7-week cycle focuses on a small, well-defined part of a larger system—and over time, these cycles build toward a complete, real-world project Feel free to join at any week and we will catch you up. There is no need to be at the beginning of the cohort to build. Complete beginners welcomed! By the End of a Full Cycle, You should Have A working project in your own GitHub repo A small but real portfolio piece you can demo or write about **Current Project** This cycle’s shared project is an **image-based knowledge extraction system**. In **Week 6**, we will finalizing and polishing your flashcard output: * Cleaning up ur pipeline and making it runnable end-to-end * Structuring your GitHub repository properly * Writing a clear README * Preparing a short project presentation/demo * Thinking about how to explain your technical decisions \*\*\* ** Who Is This For?** Beginners who learn best by doing — no prior experience needed Career switchers from product, engineering, design, marketing, or other fields who want to break into data Analysts, engineers, and PMs looking to sharpen their skills or build real-world projects Data professionals who want a creative playground outside of work Anyone with big questions and a love for learning through experimentation \*\*\* **BUILD. LEARN. EXPERIMENT. OVER COFFEE.** Come hang out, meet like-minded people, and work on something fun. Based in Berlin – in-person meetups only (for now!) Laptops + ☕ highly encouraged Join our Discord → [https://discord.gg/dPnmzcCP8w](https://discord.gg/dPnmzcCP8w) Check out our program resources: [Google Drive](https://drive.google.com/drive/folders/15DmZqD4bLBk2vU4zpvh1HiB3_5n09v-q?usp=sharing) \*\*\* ✨ Who’s Hosting? I’m Lindsey, a senior data scientist working on AI, causal inference, and data products. I’ve built models for fraud detection, uplift modeling, and LLM-based systems. I am always open to suggestion for different format for this meetup.

This edition of **Global AI Berlin** focuses on one of the most important building blocks of modern AI systems: **AI Memory**. **AI Memory and Founders Night** is an in-person AI meetup in Berlin for **developers, founders, and product leaders** working on **agentic AI systems, AI agents, and production-ready AI architectures**. The event explores how **AI memory**, **knowledge graphs**, **context management**, and **multi-agent systems** are designed and operated in real-world AI applications. The focus is on **practical, production-grade patterns**, not theory. With insights inspired by the Berlin AI ecosystem and organizations such as **Neo4j**, **Google**, **Microsoft**, and **Amazon Web Services**, the talks highlight how teams build **scalable, memory-enabled AI systems** in practice. \*\*\* \#\#\# Registration \(Limited Spots\) Primary registration via: [https://globalai.community/chapters/berlin/events/ai-memory-and-founders-night/](https://globalai.community/chapters/berlin/events/ai-memory-and-founders-night/) \*\*\* \#\#\# Speakers & Topics **Louis Buys** – CEO, Delta Campus *Incubating the Next Generation of Developers and Startup Founders* **Andreas Kollegger** – GenAI Lead, Neo4j *Context Graphs for Agents – What, How, and Why to Act* **Vasilije Markovic** – CEO, Cognee *Building Memory for AI Agents* **Elizaveta Zinovyeva** – Applied Scientist, AWS *Analyzing Codebases with Multi-Agent Systems on Amazon Bedrock* **Christian Kniep** – Agentic Thinker, MemVerge *MemMachine: Agents That Learn, Memory That Lasts* **Abdel Sghiouar** – Developer Advocate, Google *Taming Agentic AI: Running Untrusted Code Safely on Kubernetes* **Christian Glessner** – Microsoft MVP *Microsoft Foundry and Foundry IQ* \*\*\* \#\#\# Who Should Attend * AI developers and software engineers * Technical founders and startup teams * Product managers and AI architects * Anyone building **agentic AI**, **AI agents**, **knowledge graphs**, or **memory-driven AI systems**

This edition of **Global AI Berlin** focuses on one of the most important building blocks of modern AI systems: **AI Memory**. **AI Memory and Founders Night** brings together founders, engineers, and product leaders to explore how memory is designed, structured, and operated in real-world AI and agentic systems. The focus is on **production-ready patterns**, not theory. With insights inspired by work across the Berlin AI ecosystem and organizations such as **Neo4j**, **Google**, **Microsoft**, and **Amazon Web Services**, the talks cover practical approaches to building memory-enabled applications at scale. ### Registration (Limited Spots) Primary registration via: [https://globalai.community/chapters/berlin/events/ai-memory-and-founders-night/](https://globalai.community/chapters/berlin/events/ai-memory-and-founders-night/) *** ### Speakers & Topics * **Louis Buys, CEO, Delta Campus** *Incubating the Next Generation of Developers and Startup Founders* * **Andreas Kollegger, GenAI Lead Neo4j** *Context Graphs for Agents – What, How, and Why to Act* * **Vasilije Markovic. CEO, Cognee** *Building Memory for AI Agents* * **Elizaveta Zinovyeva, Applied Scientist, AWS** *Analyzing Codebases with Multi-Agent Systems on Amazon Bedrock* * **Christian Kniep, Agentic Thinker MemVerge** *MemMachine: Agents That Learn, Memory That Lasts* * **Abdel Sghiouar, Developer Advocate, Google** *Taming Agentic AI: Running Untrusted Code Safely on Kubernetes* * **Christian Glessner, Microsoft MVP** *Microsoft Foundry and Foundry IQ* *** ### Who Should Attend Developers, technical founders, product managers, architects, and teams interested in building **agentic, memory-driven AI systems**.

Remembering Stillness: A Ceremony for Deep Rest A space for woman to come into the present moment & their natural cycles with nothing to do. No expectations, no effort, just the allowance to be held in stillness and rest at the end of the week. Process: \~guided meditation \~rest & stillness -journaling \~sharing round When: Friday 25.07.2025 @7 PM-8:30PM Where: Neckarstr.12 12053 Berlin, Neukölln Doorbell: Buddhismus Berlin Payment: • Donation Based: minimum 15€ cash at entry • minimum 15€ Eventbrite Ticket: https://www.eventbrite.de/e/1440832682339?aff=oddtdtcreator • Urban Sports: https://www.urbansportsclub.com/venues/30091.

Let’s kick things off for another Meetup, this time focusing on Agentic Deployment and Flink’s Data Platform Ecosystem. Join us on February 26th in Berlin and bring all your questions! : ) **Pawel Wiszniewski: *"From Ingestion to Insight: Flink’s Data Platform Ecosystem"*** [Pawel Wiszniewski](https://www.linkedin.com/in/pawel-wiszniewski/) is a Senior Data Engineer at[ Flink](https://www.goflink.com/shop/de-DE/deals/?campaignid=23008119587&adgroupid=190905032608&creative=676430819949&utm_source=google&utm_medium=search&gad_source=1&gad_campaignid=23008119587&gbraid=0AAAAABiSYNQr0V_xYSzvjjStCH6G-hLqP&gclid=CjwKCAiAkbbMBhB2EiwANbxtbQ-a9DxijUVV0nw5nXTXr037lYGiy6MIv5YbFjeo1qsZJStEPdBNxxoCPFQQAvD_BwE) and a well-known face at our Data Engineering Meetups. We're super excited to see our guest now stepping into the presenter's role. About his keynote: "You have data, analysts, and business use cases. But you can't simply stitch them together with spreadsheets, SQL queries, and Python scripts and still expect business users to trust the dashboards. To ensure data quality, reliability, security and freshness, you need a well organised set of tools and processes: a Data Platform. In this presentation, I will explain how the Data Platform at Flink was built, outline its high-level architecture, and describe the key products, tools, and processes that enable us to operate over a thousand data models and maintain full visibility into the current state of the business." **Sneha Nafade: *"AI Agents & Agentic Deployment"*** [Sneha Nafade](https://www.linkedin.com/in/sneha-nafade/) is a Cloud Architect at [Diconium](https://diconium.com/en/?utm_term=&utm_campaign=2025+%7C+Remarketing+%7C+DE/EN+%7C+All+Visitors&utm_source=adwords&utm_medium=ppc&hsa_acc=9083156072&hsa_cam=23281606208&hsa_grp=&hsa_ad=&hsa_src=x&hsa_tgt=&hsa_kw=&hsa_mt=&hsa_net=adwords&hsa_ver=3&gad_source=1&gad_campaignid=23483494785&gbraid=0AAAAADQqBJ6jY8TgZt7Ocjy37F5Gup62I&gclid=CjwKCAiAkbbMBhB2EiwANbxtbfuZyXSVyFENwFuHv9tAcygFWNM24hiXlk3rvKVpwBFfM__jFEBHBRoCMpsQAvD_BwE) with a rich experience in Cloud & Cybersecurity Architecture & Engineering. About her keynote: "AI agents are becoming autonomous digital operators capable of managing pipelines, calling APIs, and responding to live data signals. This session provides a practical guide to designing, deploying, and governing agentic systems in real-world production environments." ✧ ✧ ✧ **What to expect:** * Two expert talks and Q&A * Networking opportunities in our great Creator Space * Some snacks & drinks to fuel your thoughts : ) ✧ ✧ ✧ **Timetable:** * 18:30 - Event admission * 18:50 - Welcome & Introduction * 19:00 - Pawel Wiszniewski: *"From Ingestion to Insight: Flink’s Data Platform Ecosystem"* * 19:30 - 5 minutes break * 19:35 - Sneha Nafade: *"AI Agents & Agentic Deployment"* * 20:05 - Snacks, Drinks & Networking * 21:30 - End ✧ ✧ ✧ More on the **-> [applydata data engineering meetup page](https://applydata.io/data-engineering-meetup/)**. **Our goal is to form a local data-loving community, so join us and let's talk data together!** ✧ ✧ ✧ *At the event, sound, image and video recordings are created and published for documentation purposes as well as for the presentation of the event in publicly accessible media, on websites and blogs and for presentation on social media. By participating the event, the participant implicitly consents to the aforementioned photo and/or video recordings. Find [more information on data protection here](https://applydata.io/events/information-on-data-protection/).*

Registration is handled by Global AI (limited spots): https://globalai.community/chapters/berlin/events/ai-memory-and-founders-night/ GDG Cloud Berlin is supporting Global AI Berlin as a community partner to amplify this event. This edition of Global AI Berlin focuses on one of the most critical building blocks of modern AI systems: AI Memory and its role in shaping intelligent applications. You’ll hear from founders, product leaders, and engineers on how memory is designed, structured, and operationalized in real-world AI and agentic systems—covering practical approaches to building memory-enabled applications at production scale. Topics include: graph-based context, retrieval layers, user interaction history, system architecture, performance considerations, and governance—plus lessons learned and trade-offs when moving from prototypes to reliable, memory-driven products. Who should attend: developers, technical founders, product managers, architects, and teams interested in using AI memory to improve application intelligence, user experience, and product differentiation. Speakers / agenda Abdel Sghiouar (Developer Advocate, Google) — Taming Agentic AI: How to Run Rogue Code Safely on Kubernetes (Google) Louis Buys (CEO, Delta Campus) — Incubating the Next Generation of Developers and Startup Founders Andreas Kollegger (GenAI Lead, Neo4j) — Context Graphs for Agents – What, How, and Why to Act Vasilije Markovic (Founder, Cognee) — Building Memory for AI Agents Elizaveta Zinovyeva (Applied Scientist, AWS) — Analyze codebase with multi-agent systems built on Strands Agents, Amazon Bedrock AgentCore (AWS) Christian Kniep (Agentic Thinker, MemVerge) — MemMachine: Agents That Learn, Memory That Lasts Christian Glessner & Luise Freese (Microsoft MVP) — Building AI Agents with Microsoft Foundry (Microsoft) Sujay Joshy (AI Engineer and strategist, Riverty) — Graph-Native Memory for Code Understanding at Scale (Riverty) Agenda *** Speaker Abdel SGHIOUAR - Google (Senior Cloud Engineer @Google Cloud) Senior Cloud Engineer @Google. GCP Certified almost everything. I tweet about GCP, Cloud, DevOps, kubernetes, and Serverless. Abdel Linkedin Profile: https://www.linkedin.com/in/sabdelfettah/; Hosted By Yevgen Batovskyi, Organizer Esra Kadah, Organizer Senior App Developer \| Author of "Flutter Design Patterns & Best Practices" \| Co\-Organizer of Flutteristas\, Flutter Berlin\, GDG Cloud Berlin\, Flutter Community AI Circle \| ‍ Antigua Mobile Hany Elfouly, Co-founder & CEO Mohamed Islem Ayari, Organizer Complete your event RSVP here: https://gdg.community.dev/events/details/google-gdg-cloud-berlin-presents-ai-memory-amp-founders-night-community-partner-global-ai-berlin/.

**What if your AI agent could decide for itself when to call a tool, or ask another agent?** Welcome to **Sprint #3 of the Agentic AI Engineering Lab**, where we level up from building protocols to building *decision-making agents*. In this hands-on lab, you’ll create an autonomous agent that: * speaks the **A2A protocol** to collaborate with other agents * registers to an agent hub for discovery * connects to an **MCP server** when tools are needed * intelligently chooses ***when to delegate vs when to execute*** No lectures. Just systems thinking + real building. *** ### What We’ll Do * **10:00** \- Quick round of introductions * **10:15** \- Recap: AI Agents\, MCP Protocol and A2A Protocol * **10:30** \- Build your decision\-making agent * **12:00** \- Live demos \+ technical teardown * **12:30** \- Wrap\-up \+ takeaways *** ### What to Expect * Hands-on building, not passive listening * Real agent-to-agent experiments * Collaboration with fellow builders * Fast ideas, quick tests, lots of learning * Unexpected behavior (the fun part) ### What Not to Expect * Long lectures * Step-by-step tutorials * Sales pitches * Perfect code This is a **lab**; we learn by building, breaking, and iterating. *** ### ‍ Who Should Join Curious developers, tinkerers, and engineers, with basic knowledge of AI Agents. **Bring:** * Laptop with dev setup * Access to an LLM (Claude / OpenAI / Ollama) * Comfort building small things in any language *** ###

**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** *Spec-Driven Development with GitHub Spec-Kit: From Intent to Implementation* Spec-driven development flips the traditional workflow on its head: instead of code being the source of truth, the specification becomes the backbone of design, collaboration, and delivery. In this session, we’ll explore how GitHub Spec-Kit enables teams to treat specifications as first-class artifacts—living documents that drive architecture, implementation, and verification. You’ll learn how Spec-Kit helps teams clearly express intent using structured, version-controlled specs that live alongside code. We’ll walk through a practical workflow that starts with defining system behavior and constraints, then progressively refines those specs into testable, automatable outcomes. Along the way, we’ll show how specs can reduce ambiguity, improve cross-functional collaboration, and make design decisions explicit before a single line of production code is written. This talk will cover: --What spec-driven development is (and what it isn’t) --How GitHub Spec-Kit fits into modern developer workflows --Using specs to align product, engineering, and AI-assisted development --Real-world examples of turning specs into implementations with confidence Whether you’re building greenfield systems, integrating AI into your stack, or trying to reduce costly rework, spec-driven development offers a scalable way to move faster without sacrificing clarity. Attendees will leave with concrete patterns and a clear mental model for adopting GitHub Spec-Kit in their own projects. **YouTube Link** TBA

If you enjoy knitting, crocheting, working with yarn, this is the group for you. Meet new friends, bring your projects, learn from others.
We normally meet a few times a month during the week from 6-8pm. We will occasionally meet on weekends during daytime hours.

We will show a variety of methods for building agents that run in Microsoft Foundry. This covers the different types of agents: Prompt, Multi, and Hosted, as well as the development lifecycle using evals and traces.

## Overview Networking event for aspiring school leaders interested in building their own high-quality K-12 public school in their community. **About the Event:** Aspiring school founders in the Columbus, OH, area are invited to join us for this special in-person networking event where you will be able to ask questions, learn more about the BES Fellowship and make sure you are prepared for the application process and beyond. In addition to meeting like-minded leaders who are also committed to transforming education and bringing a high-quality K-12 option to their community, you’ll have the opportunity to meet our team and gain special insights from Dr. Chaundria Smith, BES Senior Director of Fellowship Coaching, who works closely to support BES Fellows. During this event, you can also learn more about the various regions across Ohio we are prioritizing for the BES Fellowship to ensure that new schools will respond to the needs of their community. **About the BES Fellowship:** Leaders in the BES Fellowship embark on a multi-year process to design, found, and lead an excellent K-12 public school that reflects and responds to community needs. We are currently accepting applications to the 2026 BES Fellowship. The extended deadline is March 30, and the 2026 Cohort will begin August 2026. Applications will be reviewed on a rolling basis until all spots are filled. Learn more at [bes.org/fellowship](http://bes.org/fellowship "http://bes.org/fellowship").

Meeting @ Improving's office in Downtown Columbus, near Cosi. Parking is paid, food & drink are free! This time we're hanging out after the long winter break. We are going to look at a group-member's project used to help learn the fundamentals of Go, and if time allows, we can take a look at how we like to build Go projects using AI.

Join us and work remotely together with like-minded UX folks. This is a last minute opportunity to get out at one of our favorites, [Junto](https://thejuntohotel.com/)! Plenty of seating is available in addition to a connected coffee shop and restaurant. I will be there 10-12 for sure, maybe longer. Feel free to come and go anytime during the day, and don't forget your charger and headphones!