MLOps Community Amsterdam Meetup @ JetBrains

Welcome to a new live get together for the global MLOps.community in Amsterdam. Together with our host, JetBrains, we will enjoy a series of talks and ample time to socialize with others in the community!

Schedule:
18:00 — Doors open
18:30 — 🎤 JetBrains Talks (2x 20mins)
19:10 — 🎤 What does it take to add Copilot to Obsidian by Jordi Smith (ML Engineer @ Xebia)
19:45— 🥤 Drinks & Networking

Sign-up instructions:

• Sign up via meetup.com
• Let us know if you have any strict dietary restrictions (e.g. vegan🌱)
• We are looking for speakers for the next events. If you would like to give a talk, let us know the topic and a contact information.

🎤 Talks

AI Assistant: Deterministic code analysis meets AI creativity.
JetBrains tools understand the code of your projects using different code analysis functionality. This year, our team worked on joining it with AI intuition. As a result, we are happy to present the AI Assistant plugin for IntelliJ IDEA. I will make an overview of the AI Assistant plugin features and the technical aspects of its implementation: prompt context gathering, IntelliJ IDEA project indexes in Retrieval Augmented Generation (RAG), and plugin features evaluation.
By JetBrains

Do You Need to Adopt a Pile of MLOps Tools, or Don't You?
MLOps practitioners frequently face the challenge of choosing which tools to integrate into their processes. We investigate the current methods for executing ML experiments, identifying their disadvantages and exploring which tools can remedy these issues. We argue that most of an ML engineer's needs can be met using an IDE, providing a user-friendly alternative. In conclusion, we highlight how a lean selection of tools can proficiently accomplish all tasks for an ML engineer, furthermore offering a practical countermeasure to the persistent GPU shortage issues.
By JetBrains

What does it take to add Copilot to Obsidian?
Ever since I got access to GitHub Copilot, I have been truly amazed by its capabilities. It continuously keeps feeding me possible completions for my code and text. They might not always be perfect, but they are often good enough to be used as a starting point and prevent me from suffering from the white page syndrome. I'm also an avid user of Obsidian, a note-taking application, where I often encounter the same white page syndrome. This often resulted in me either writing my longer notes in my IDE using Copilot or procrastinating and not writing the notes at all. The engineer in me saw this as a challenge, which resulted in the Obsidian Copilot plugin. In this talk, I will discuss this plugin and its inner workings and design. I will answer questions such as:

• How do we obtain completions that take both the text before and after the cursor into account?
• How do we ensure that the obtained completions are the type of completions we want?
• What kind of pre and post-processing is needed for these completions?
• How do we design a maintainable software architecture for such a plugin?

by Jordi Smit, Machine Learning Engineer @ Xebia Data