NovembAI: Agents, Algorithms & Audio

Join us for an inspiring evening of AI, software development and machine learning at the Evidi office in Stockholm.

We’ll bring together three MVPs for three very different sessions, covering everything from working with GitHub Copilot Squad and understanding what really happens inside coding agents, to exploring the algorithms and machine learning models behind modern music technology.

Expect practical demos, technical deep dives and plenty of opportunities to exchange ideas with fellow developers. Food and beverages will be served after the first session, with breaks between the talks for networking and conversation.

All sessions will be held in English.

📅 24 November, doors open 17:15, sessions start 17:30.
📍 Evidi, Kammakargatan 9A, 111 40, Stockholm
🍽️ Food will be served

Sessions:

Tore Nestenius - How Coding Agents Actually Work: Tools, MCP, and the Agent Loop
.NET MVP, TN DataKonsult AB
Coding agents read files, run commands and change entire codebases. Yet the language model at their core only takes text in and sends text out. So what is actually happening under the hood?

This talk opens the black box and focuses on mechanics rather than prompting techniques. How does the agent talk to the model? How do tools give it access to source code, and how does MCP fit into the same tool model? What really goes into each request as the agent loop runs?
Finally, we put the pieces together and build a basic coding agent in C#. The goal is mechanical sympathy for agentic development: understanding how it works, not just how to use it.

Martin Ehrnst - Assemble! coding with a Squad
Azure MVP, Evidi
AI-assisted development and cloud engineering is moving beyond prompt and assistant. In this session we will explore GitHub Copilot Squad, how it works and what this way of working can mean for your everyday workflows.

Alan Smith - Music and Machine Learning
AI MVP, Evidi
In the 70s, pioneers like Kraftwerk and Brian Eno began experimenting with computers in music. Today, the laptop is the recording studio for many musicians. Behind its user interfaces lies a hidden world of algorithms and models, from old-school Fourier transforms to state-of-the-art CNNs and transformers.

In this demo-intensive session, Alan lifts the lid on digital signal processing and machine learning for music analysis. Using custom-trained and open-source models, he shows how software can classify a genre, find the beat, separate the vocals and even read the mood of the lyrics. He also explains the theory that makes it work.

If you have an interest in music, algorithms, local models or creative programming, or you just want a session that's not all about agents, this is the one for you!

Schedule:

17:15–17:30 Welcome and introduction
17:30–18:15 Tore Nestenius: How Coding Agents Actually Work: Tools, MCP, and the Agent Loop
18:15–18:35 Food and beverages
18:35–19:05 Martin Ehrnst: Assemble! Coding with a Squad
19:05-19:15 Break
19:15–20:00 Alan Smith: Music and Machine Learning
20:00-20:30 Networking and discussions

Food and beverages will be served.
Do you have any allergies or other preferences? Please contact klara.svanstrom@evidi.com.

RSVP
Please RVSP before 12:00 Monday the 23rd November.

We look forward to an inspiring night together in Stockholm!