From Data to Deployment: Harnessing Generative AI


Details
Another month, another meetup! 🤖
Join us for another exciting meetup this month at the AI Sweden office! AI Sweden will dive into discussions about a Scandinavian multimodal pre-training dataset for their next-generation foundation model. Following this, Sana Labs will provide insights into building products and technology using LLMs and generative AI.
Agenda:
17:30 - 18:00: Doors open
18:00 - 18:10: Welcome
18:10 - 18:40: Data curation at scale for foundation model pre-training
18:40 - 19:10: Break
19:10 - 19:40: Sana AI - How to build a consumer-grade enterprise product with Generative AI
19:40 - 20:30: Networking
–
Presentations:
Data curation at scale for foundation model pre-training
Tobias Norlund - Research Scientist, AI Sweden
The capabilities of large foundation models are fundamentally formed by the data they are trained on. This data is the single most defining feature of the resulting model's behavior, and can be a differentiator to all other models out there. In this presentation, I will present AI Sweden's ongoing work of curating a unique Scandinavian multimodal pre-training dataset for our next generation foundation model. We will in particular focus on our pipeline for extracting high quality data from large web dumps (common crawl) and compare to existing datasets.
Speakers Bio:
Tobias Norlund is a Research Scientist at AI Sweden, currently working on developing next generation Swedish and Scandinavian foundation models. He holds a degree of Licentiate in NLP from Chalmers University and 8+ years of experience in AI-based product development from Schibsted and Recorded Future.
Sana AI - How to build a consumer-grade enterprise product with Generative AI
Jacob Jonsson - Senior Product Manager, Sana Labs
Sana AI is at the forefront of helping people use the knowledge within their organizations. Developed by Sana Labs, a Stockholm-based company known for its work in AI for learning and knowledge, Sana AI is designed to help people work smarter and more efficiently. It's built on the latest large language models (LLMs) and a unique retrieval algorithm. With just a few clicks, you're up and running, fully integrated with your entire knowledge base.
This talk will dive into the technology that powers Sana AI. A key aspect of its success is the use of Retrieval Augmented Generation (RAG). Sana AI's retrieval algorithm does more than just search through files from different apps; it can also transcribe meetings, understand what's being discussed, and summarize the key points. It covers everything from effective indexing of files and meeting content to sophisticated query planning and source attribution. This ensures that users can quickly find precise, relevant information, whether it's buried in documents or was discussed in meetings. We'll also explore how to think about building products and technology using LLMs and generative AI. The goal is to provide insights into the principles of designing AI-driven tools that are not only powerful but also intuitive and frictionless.
Speakers Bio:
Jacob Jonsson is a Senior Product Manager at Sana Labs, leading the product team building Sana AI. Jacob has prior experience driving AI transformations from the Boston Consulting Group. With a background in Engineering Physics & Machine Learning, Jacob has a passion for all parts of AI — from making to taking to market.
–
About the event:
Date: April 18th, 17:30 - 20:30
Location: AI Sweden Office (Folkungagatan 44, 118 26 Stockholm)
Directions: 1-minute walk from the Medborgarplatsen station.
Tickets: Sign up required. Anyone who is not on the list will not get in. The event is free of charge.
Capacity: Space is limited to 60 participants. If you are signed up but unable to attend, please change your RSVP by April 17th.
Questions: Please contact the meetup organizers.
–
Code of Conduct
The NumFOCUS Code of Conduct applies to this event; please familiarize yourself with it before attending. If you have any questions or concerns regarding the Code of Conduct, please contact the organizers.
COVID-19 safety measures

From Data to Deployment: Harnessing Generative AI