Cloud Infrastructure

**A software engineers guide to LLMs** In this talk Richard Conway will break down what LLMs are layer by layer, how they work, how they use software principles to work and how to use them in software. This will be a fluid talk with a lot of philosophy, maths from school and how to build your software understanding things like hallucinations and new techniques that can make your management of LLMs more deterministic. We'll consider a lot of things here including the new code generation tools which have developers quivering in their boots, what they mean for our industry and how to use them for effect. Lots of mini-demos and great discussions, good people, beer and pizza! **About the speaker** Richard Conway co-founded the user group, along with Andy Cross, Joe Hancock and Chris Parsons. He is a Microsoft AI MVP and a Microsoft Regional Director and has been writing software in one form or another since his fingers were large enough to hit a ZX81 keyboard.

We'll be meeting to play games. Invitation open to all members and drop ins. No need to bring your own games as there are always options from other members but please do so if there's something you're interested in playing. Some longer games we've played recently include Eclipse, Dungeon Lord, Hansa Teutonica, Star Wars: Outer Rim and Inis. But we've also played medium/light games like Heat, Flamme Rouge, Fallout and Ride the Rails. Plus some short card games like Lords of Scotland and Skull King. Most people generally arrive by about 7:30. You can park at the pub for free if you validate at the bar and there may also be space on nearby streets. Please also see our facebook group for some further details and discussion: https://www.facebook.com/groups/315820376735888 Attendance is free and the pub's bar and kitchen should be open.

Registration Portal: [Event Registration Portal](https://luma.com/cclondon26?utm_source=sherif) About Event ​"Why are there no coffeeshops open late? But what if I want to co-work at night?!?" - everyone on Twitter ​Cerebras is excited to present, Cafe Compute. The first, late-night pop-up coffeeshop with a barista bar ☕️, cozy snacks and couches 🛋️, and more. ​​​​​​This event is brought to you in collaboration with OpenAI. ​​Cerebras is the world’s fastest AI inference, up to 15x faster than leading GPUs. Cerebras Inference is powered by our Wafer-Scale Engine (WSE-3) - the world's largest AI chip. Experience the speed for yourself with OpenAI Codex-Spark, powered on Cerebras, and get free compute at cerebras.ai. Location Please register to see the exact location of this event.

**Venue:** Riverbank House, 2 Swan Ln, London EC4R 3AD **Please note:** 1\. 🚨🚨🚨 A valid photo ID is required by building security\. 🚨🚨🚨 2\. This event follows the [NumFOCUS Code of Conduct](https://numfocus.org/code-of-conduct). Please familiarise yourself with it before attending. If your RSVP status says "You're going" you will be able to get in. No need to show your RSVP confirmation when signing in. If you can no longer make it, please unRSVP as soon as possible. **Code of Conduct:** This event follows the NumFOCUS Code of Conduct. Please get in touch with the organisers with any questions or concerns. As always, there will be free food and drinks, generously provided by our host, Man Group. **Main Talks** 1. **[Nicoleta Lazar](https://www.linkedin.com/in/nicoleta-lazar-921a6864/) Query federation in modern OLAP databases** As analytics ecosystems grow more diverse, organisations increasingly need to query data across warehouses, data lakes, and operational systems without excessive data movement or duplication. **Query federation** has become essential by enabling unified SQL access and intelligent predicate pushdown into heterogeneous sources. This talk introduces the core principles of federation and why it matters for modern OLAP workloads. Using StarRocks as a model system, we highlight its vectorised execution engine, native connectors, and deep Apache Iceberg integration that together deliver high-performance lakehouse querying. 2. [Ben Guerin](https://www.linkedin.com/in/bjhguerin/) **How I Learned to Stop Worrying and Let Claude Code Write the Python [the story of [ismypubfucked.com](http://ismypubfucked.com/ "http://ismypubfucked.com")]** 43,000 pubs. Official VOA data. One question. [ismypubfucked.com](http://ismypubfucked.com/ "http://ismypubfucked.com") got 400k visits in two weeks and wall-to-wall media coverage: from the Mirror to the Telegraph to City AM. I'm not a Python developer. I used Claude Code to wrangle government spreadsheets, match them to OpenStreetMap, and ship a working site. All in less than 6 hours from initial idea to a live website. The AI wrote the Python, I just knew what story the data could tell. This talk covers what happens when tools like Claude Code unlock Python for people who think in narratives rather than syntax, and what a decade of building viral campaigns has taught me about making people actually give a shit about spreadsheets. **Lightning Talks** 1. **TBC** 2. **TBC** \-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\- **Logistics** Doors open at **6.30 pm** (get there early as you'll need to sign in with building security). Talks start at **7:00 pm**, with drinks afterwards from **9:00 pm** at The Banker (EC4). We have reduced capacity for this event, but there will be plenty of people to discuss data science questions with. Please unRSVP in good time if you realise you can't make it. We're limited by building security on the number of attendees, so please free up your place for your fellow community members. If you want me to trim lightning talks down to two or shorten any abstracts, say which ones.

April Edition ✨✨! We will be at Kraken Tech (Octopus Energy Group), London office! (in-person only event, no streaming) **Talks:** **Digitising Historical Caving Data with Django** - *Andrew Northall* Building an automated pipeline with Django which can extract, structure and publish 2,700+ historical printed reports from scanned PDF copies. *Andrew Northall* is a Python & Django specialist who maintains several open source Django projects/services for the caving and amateur radio communities. **llms.txt for Django: what works** - *Thibaud Colas* The llms.txt format is an emerging standard to structure information for Large Language Models. It’s a desirable addition to the docs of Python packages. Let’s review how to adopt it! We’ll discuss the fundamentals of the format and its benefits as a user of the docs, and as a maintainer. How to produce and consume those files across different tools (Sphinx, mkdocs, Django). How to optimize them for different LLMs with an eval suite. Tools and techniques you should be able to reuse through other engineering tasks with LLMs. *Thibaud Colas* works at Torchbox as developer, core contributor to the Wagtail CMS build on Django. **Want to share your Django or Python knowledge?** We’d love to feature your talk! Just fill out our [speaker form](https://forms.gle/LkaRdKVTF9ApbrEq7) to propose a talk! **Agenda:** • 6:15pm Doors open, socialising • 6:30pm Food and drinks offered by our sponsor • 7:00pm Doors close \*\* • 7:15pm Introduction, News, Talks • 8:30pm Socialising • 9:30pm Fin **Please make sure to arrive between 6:15 PM and 6:55 PM**. If you arrive earlier than 6:15 PM or later than 6:55 PM, you can take a seat in the seating area located in the downstairs reception. **Please note that we cannot accommodate arrivals after 7:15 PM.** Our meetups are governed by a [Code of Conduct](https://www.djangolondon.com/code-of-conduct/). Please take a few minutes to read it. This meetup is sponsored by: • [Kraken Tech (Octopus Energy Group)](https://kraken.tech/): "Kraken is a revolutionary energy tech platform built with predominantly Python & Django. We Serve 54 million households worldwide and have 38GW of contracted capacity." • [JetBrains](https://www.jetbrains.com/): "We make professional software development a more productive and enjoyable experience." Website: [https://djangolondon.com](https://djangolondon.com/) BlueSky: [@djangolondon.com](https://bsky.app/profile/djangolondon.com) Github: [github.com/djangolondon](https://github.com/djangolondon/) Open Collective: [opencollective.com/the-london-django-meetup-group](https://opencollective.com/the-london-django-meetup-group/)

**Important:** Register on the [event website](https://www.aicamp.ai/event/eventdetails/W2026040809) is required for admission. **RSVP on meetup is turned off** Enjoy afterwork drinks and practical discussions on Apache Kafka, ClickHouse, PostgreSQL and much more. Enjoy a session designed for data engineers, architects and technical decisions-makers like you, who want to address the challenges they face in data-infrastructure. No high-level pitches but straight into practical discussions on how to build scalable, efficient open sourced-powered data platforms. **Why Attend?** * Connect with local data and tech professionals * Get practical insights into streaming, storage, and analytics with Kafka, Iceberg, PostgreSQL, and ClickHouse * Learn how to reduce operational overhead while scaling your data infrastructure performance * Bring your questions and challenges for direct discussions with Aiven experts **Agenda:** * 4:00pm\~4:30pm: Checkin, Food, Networking * 4:30pm\~6:00pm: Tech talks and Q&A * 6:00pm\~7:00pm: Happy hour, Open discussion, Networking * **Tech Talk: Context Is the New Infrastructure** **Speaker:** Stan Dmitriev (Aiven) **Tech Talk: Apache Kafka on Aiven** **Speaker:** Hugh Evans (Aiven)

Meet out front at ten to 1pm (12:50) and go in together Join Lester Hillman for a talk about John Harrison, master clockmaker. I have long been fascinated by historic issues with measuring longitude. Rather important for ships to know where they had sailed to! This problem couldn't be solved until clocks were accurate enough, and capable of functioning aboard ships. Please join me. RSVP as soon as you know, I still need to request more tickets. The attendance fee is £2.50, thank you.

We are back! This is going to be a monthly IN PERSON meet up in London. A very big thank you goes to Accurx who is hosting us in their amazing Shoreditch offices. This will include access to facilities including coffee! 😻☕️ **Want to know more? 👇** As with all CoffeeOps, this is an interactive, chatty, build your own adventure type of meet-up. Instead of booking in experts to *learn* *from*, we gather a group of practitioners to ***learn*** ***with***. We have been running for over 4 years now and often get feedback that this is the most useful meetup people join! The event is approximately 90 minutes and always starts by brainstorming topics to discuss. The group then votes on the most relevant/interesting to them and we then cycle through the topics trying to discuss each deeply, but in a time bound slot. To get an idea of topics, you can check out our remote board [here](https://easyretro.io/publicboard/h404NF84Uje4tCcxpJdHr4Ycs9B3/a4a35d2b-181b-409b-832e-a729be1af79c). The general gist is anything is fair game, though we tend to revolve around software delivery and definitely lean towards the more Ops-y or DevOps-y side of things.

The Databricks London Meetup is back after a short break, and we’re looking forward to bringing the community together again for another friendly, in-person evening of data, AI, and good conversation. Join us for a relaxed meetup where data and AI professionals can swap ideas, hear a couple of practical talks, and meet others working with Databricks in the real world. Whether you’re deep into the platform already or just keen to learn from the community, this is a great chance to pick up something useful and meet some excellent people along the way. This session will include the latest Databricks updates, a practical talk from the Advancing Analytics team on how they used Lakebase to power AI projects, and time for questions with the Databricks team before we round things off with pizza and networking. **17:30 - 18:00:** Arrival & Networking **18:00 - 18:10:** Opening Remarks & Introductions **18:10 - 18:30** What’s New in Databricks - Simon Whiteley **18:30 - 19:00** Building Better AI Projects with *Lakebase* \- Toyosi Babayeju from Advancing Analytics **19:00 - 19:15** Q&A with the Databricks team **19:15 onwards:** Pizza, Drinks & Networking *Join us for a fantastic evening of learning and networking at the London Databricks meetup!*

Elixir London Meetup; February 2026 edition. All are welcome. **Session 1: Anton Borisov - Distilling Kafka's Binary Protocol into Elixir** We'll trace a produce request to show what any Kafka client has to get right. In the BEAM ecosystem, erlkaf wraps librdkafka, brod implements the protocol in Erlang, and [kafka_ex](https://hex.pm/packages/kafka_ex) goes pure Elixir. Kafka_ex's approach is to generate serializers at compile time from Kafka's own schema definitions: plain structs, pattern matching, structural boilerplate isolated from the logic you actually think about. That design is what let us jump two major versions ahead and adapt to significant protocol changes. Let's dig into the idea, the code, and the tradeoffs. **Session 2: Evadne Wu** **\- JavaScript Execution Strategies in Elixir\-based Agent Frameworks** We'll look at integration points between elixir agents (in the AI fashion) and JavaScript execution, and all the foot guns involved. *The event is kindly sponsored by* [Fresha](https://www.fresha.com). **Time** Doors open at 18, start at 18.30. **Venue** Fresha The Bower, 207-211, Old St, Tower, London EC1V 9NR **Refreshments** Food and drinks will be provisioned on the day. **IMPORTANT**: **If you wish to take photos and post online during the event, please notify organisers first, so that attendees can opt-out from being in photos.**

Agents are writing code, running workflows, and building other agents. ​ The challenge isn’t basic functionality. It’s getting these systems to operate predictably in production, and integrating them into how teams build and ship software. ​This session covers both. ​ From a production lens: how to build agentic systems using spec-driven development, structured context, and evaluation loops that hold up at scale. ​From a human lens: what breaks when agents become collaborators. Ownership, coordination, and how teams actually adapt. ​Speakers: ​**Clifton Cunningham** ​**Member of Technical Staff, [guidemode.dev](https://guidemode.dev/?utm_source=luma) \| Former CTO at DailyMail\, Infinitas Learning\, TES Global** ​Clifton will explore what happens when AI agents become part of your team. Drawing on experience leading large-scale engineering organisations, he’ll unpack the human, cultural, and operational challenges of working alongside AI colleagues. ​**Alfonso Graziano** ​**AI Tech Lead\, Nearform \| Author of *AI-Native Software Engineering*** ​Alfonso will walk through how to build agents that build and maintain other agents. A practical, production-focused session covering spec-driven development, context engineering, evaluation loops, and how to make agentic systems reliable at scale. ​ Join us for pizza, drinks, and practical AI insights.

The global Azure community is coming together again, and Columbus is officially on the map. View the session lineup and speakers at [Global Azure Columbus 2026](https://coazure.github.io/cbus-global-azure-2026/) On **Saturday, April 18, 2026**, the Azure Columbus Meetup, DevOps Meetup, and Code and Coffee Meetup are hosting our local edition of Global Azure 2026. This is a free, community-driven event packed with learning, networking, and all things Microsoft Azure. Whether you’re building modern cloud-native apps, experimenting with AI agents, deploying containers, automating infrastructure, or just beginning your Azure journey, this event is for you! **What to Expect** * Engaging technical sessions * Real-world Azure architecture & cloud-native patterns * AI, agents, automation, and modern DevOps * Food and drinks (because learning burns calories) * Time to connect with fellow engineers, architects, and cloud enthusiasts **Who Should Attend?** * Software engineers (any language, any stack) * Cloud architects * DevOps engineers * Data professionals * AI explorers * Platform builders * Anyone who loves solving hard problems with great tools If you build, deploy, automate, scale, monitor, or optimize in Azure, you’ll feel right at home. **Why Global Azure?** Global Azure is a worldwide community event where Azure user groups host learning sessions on the same day across the globe. It’s grassroots. It’s technical. It’s practical. And it’s powered by people who genuinely love sharing what they’ve learned. And yes, it’s free to attend!

Topic to be announced. **CALL FOR SPEAKERS** Learn more: https://www.awscolumbus.com/get-involved/ **THANK YOU** *VEEAM* for hosting our meetup! To learn more about *Veeam*, please visit their website: https://www.veeam.com/ **DIRECTIONS** 8800 Lyra Dr #450 · Columbus, OH go to 4th floor. **Want to sponsor the pizza and/or bar tab?** Please contact me if you would like to sponsor this meetup's pizza and/or bar tab: angelo@mandato.com

Topic to be announced! **CALL FOR SPEAKERS** Learn more: https://www.awscolumbus.com/get-involved/ **THANK YOU** *Leading EDJE* for hosting our meetup! Leading EDJE specializes in helping organizations innovate and solve complex problems. EDJE does this with their "A" player team of professionals using agile methodologies and SCRUM management techniques to give organizations a competitive advantage that sets them apart. Learn more at https://www.leadingedje.com/ **THANK YOU** *Amazon Web Services* for sponsoring pizza! Learn more at https://aws.amazon.com/. **DIRECTIONS** 6515 Longshore Loop, Suite 300 Dublin, OH 43017 (3rd floor) **FREE PARKING** at the Endres Garage across the street, 6540 Riverside Dr, Dublin, OH **Want to sponsor the pizza and/or bar tab for a future meetup?** Please contact me if you would like to sponsor this meetup's pizza and/or bar tab: angelo@mandato.com

The global Azure community is coming together again, and Columbus is officially on the map. View the session lineup and speakers at [Global Azure Columbus 2026](https://coazure.github.io/cbus-global-azure-2026/) On **Saturday, April 18, 2026**, the Azure Columbus Meetup, DevOps Meetup, and Code and Coffee Meetup are hosting our local edition of Global Azure 2026. This is a free, community-driven event packed with learning, networking, and all things Microsoft Azure. Whether you’re building modern cloud-native apps, experimenting with AI agents, deploying containers, automating infrastructure, or just beginning your Azure journey, this event is for you! **What to Expect** * Engaging technical sessions * Real-world Azure architecture & cloud-native patterns * AI, agents, automation, and modern DevOps * Food and drinks (because learning burns calories) * Time to connect with fellow engineers, architects, and cloud enthusiasts **Who Should Attend?** * Software engineers (any language, any stack) * Cloud architects * DevOps engineers * Data professionals * AI explorers * Platform builders * Anyone who loves solving hard problems with great tools If you build, deploy, automate, scale, monitor, or optimize in Azure, you’ll feel right at home. **Why Global Azure?** Global Azure is a worldwide community event where Azure user groups host learning sessions on the same day across the globe. It’s grassroots. It’s technical. It’s practical. And it’s powered by people who genuinely love sharing what they’ve learned. And yes, it’s free to attend!

Colombus Code & Coffee is thrilled to partner with Central Ohio Azure to put on the global Azure community day! RSVP here: https://www.meetup.com/central-ohio-azure/events/313484939/ View the session lineup and speakers at [Global Azure Columbus 2026](https://coazure.github.io/cbus-global-azure-2026/) On **Saturday, April 18, 2026**, the Azure Columbus Meetup, DevOps Meetup, and Code and Coffee Meetup are hosting our local edition of Global Azure 2026. This is a free, community-driven event packed with learning, networking, and all things Microsoft Azure. Whether you’re building modern cloud-native apps, experimenting with AI agents, deploying containers, automating infrastructure, or just beginning your Azure journey, this event is for you! **What to Expect** * Engaging technical sessions * Real-world Azure architecture & cloud-native patterns * AI, agents, automation, and modern DevOps * Food and drinks (because learning burns calories) * Time to connect with fellow engineers, architects, and cloud enthusiasts **Who Should Attend?** * Software engineers (any language, any stack) * Cloud architects * DevOps engineers * Data professionals * AI explorers * Platform builders * Anyone who loves solving hard problems with great tools If you build, deploy, automate, scale, monitor, or optimize in Azure, you’ll feel right at home. **Why Global Azure?** Global Azure is a worldwide community event where Azure user groups host learning sessions on the same day across the globe. It’s grassroots. It’s technical. It’s practical. And it’s powered by people who genuinely love sharing what they’ve learned. And yes, it’s free to attend!

Pay attention! We’re delaying a week for those of you who may be on Spring break and/or celebrating the holiday. Ok, no pancakes, but waffles-a-plenty. Join the local UX and Digital Design community for a casual monthly breakfast. Look for us upstairs!

The Fellowship of the Star Schema: Transforming OLTP Data for Power BI One does not simply build reports on OLTP data. Join us on an epic journey from the depths of raw, normalized tables to the shining halls of a well-modeled star schema fit for Power BI greatness. We will demystify the differences between OLTP and OLAP, walk through star schema vs snowflake (and why Power BI has a clear favorite), and show you the compression differences between normalized and columnar storage that will make you never look at an OLTP model the same way again. We will cover the different types of dimensions, when to use them, and how to handle Slowly Changing Dimensions without losing your mind. You will leave knowing exactly how to identify a bad model, how to convert OLTP data into a clean star schema, and how to forge relationships that rule them all. Whether you are a data wizard, a curious hobbit, or somewhere in between, your models will never be the same.