Berlin AI Meetup October Edition 🍂

Hi friends,
​
🚀 Join us on October 6 for our first Berlin AI meetup of the new season.
​This time, we’ll explore how Large Language Models and Physics-AI can be used for scientific simulations and dive into building better knowledge graphs through agent steering.

Reserve your seat here → REGISTER on Luma

🗣 Call for Proposals
If you’re building, applying, or experimenting with AI, we’d love to hear your story. Submit your talk via our CFP form and take the stage at one of our upcoming meetups.

🤝 Organized by
This event made possible thanks to the support from AI Coding Summit Berlin organizers – GitNation

🤝 Hosted by
Huge thanks to our friends from Enpal!
At EnpalTech we’re driving Enpal’s mission to revolutionize sustainable energy. Our team builds innovative applications that enhance every stage of the customer journey, from sales to energy management. With Enpal’s Energy Platform and Virtual Power Plant, we integrate decentralized energy sources to create a smarter, more efficient grid, accelerating the transition to renewable energy. We deliver impactful solutions — from operational software in Enablement Platform to appointment management in CraftOS — and stay agile to ensure renewable energy is accessible to all. Software Engineering job page: Enpal Teams: Software Engineering
​
🕑 Event Schedule
18:30 - Doors open
19:00 - Intro and announcements
19:10 - Large Language Models and Physics-AI for Scientific Simulations - A Seshaditya
19:35 - From Copilot to Agentic Engineering: How AI Agents Are Redesigning Software Development - Oleksii Rak
20:00 - Break
20:15 - Better Knowledge Graphs Through Agent Steering - Evgenii Demidov and Aakriti Jain
20:40 - Open Demo Time - show your project, open source work or a small demo of what you learned recently
Mingle until 21:30

🗣️ Talks
➡️ From Copilot to Agentic Engineering: How AI Agents Are Redesigning Software Development - Oleksii Rak
The central idea: AI is moving from suggesting code to independently handling longer tasks—investigating bugs, changing multiple files, running tests, reviewing results, and preparing pull requests. The engineer's role shifts from primarily writing code to specifying, delegating, reviewing, and controlling parallel AI work.

➡️ Large Language Models and Physics-AI for Scientific Simulations - A Seshaditya
In this talk, we present how large language models (LLMs) and physics-inspired AI models (Nvidia PhysicsNemo) can be combined to make complex scientific simulations such as CFD (fluid dynamics) more accessible, automated, and interactive. By integrating AI agents and natural language interfaces with open-source tools such as MetaOpenFOAM, OpenFOAM-GPT, and Foam-Agent, users can configure, run, and analyze simulations using Python-centric, reproducible workflows

➡️ Better Knowledge Graphs Through Agent Steering - Evgenii Demidov and Aakriti Jain
Using agents to create knowledge graphs from unstructured documents accelerates the process but comes with many challenges. This talk presents agent steering, a method to improve issues like completeness, entity resolution, and hallucinations. You'll walk away with practical guidelines on how to engineer custom harnesses for knowledge graph creation and how to close the reliability gap with agent steering.
​
👍 Code of Conduct
By registering for this event you agree to comply with our CoC
​
📩 Contact
events@gitnation.org