Vector DB & LLM Hackathon
Join us on June 16th & 17th, 2023 in Berlin for a welcome evening of BBQ (Friday, June 16th - Session 1) and an all-day Vector Search Hackathon (Saturday, June 17th - Session 2) with the MLOps.community and Weaviate using a large corpus of community Slack and Videos!
Need a vector database brush up? Join us for a virtual workshop on Wednesday, June 14th for an intro to Weaviate, where you'll get hands on experience with a vector database.
💪 Collaborate with and learn from other like-minded MLOps practitioners in AI, ML, and Data Science industries.
💻 Get access to tools and credits from Weaviate to build end-to-end solutions.
👨🏻💻 Get support in-person from Weaviate engineers – who will help you go from 0 to 100 in no time.
💰 Chance to win cash prizes!
📣 Selected teams will be invited to share their solutions at an upcoming Berlin MLOps Community meetup
🦾 Intro:
Vector embeddings generated by LLMs – like those provided by OpenAI, Cohere, Hugging Face, and Google PaLM, or those you may fine-tune yourself – opened up a world of possibilities in how we understand and query our data.
Combine this with Weaviate – an AI-native vector database – and you have a recipe for running LLMs at a scale of millions or even billion data objects.
🦾 The Challenge(s):
Now the question (and the challenge) is:
“What can you build with a dataset + LLM + vector database in a day?”
Option 1 – MLOps community Slack search
MLOps.community is a 17K+ member community, primarily online in a thriving slack channel and through virtual and in-person events. There is a lot of tribal knowledge in the community where experts answer questions from other MLOps practitioners in Slack and via in-person meetups and online videos. We want you to build an LLM-powered application (Web/Mobile/Slack) to answer MLOps and other ML/DS/AI questions that community members may have using the historical knowledge in Slack, videos, etc. For example - you may implement the solution as a chatbot that uses embeddings and a vector datastore to generate the answers.
Option 2 – bring your own use case
Bring in your own data and a use case – that you were itching to build but never had the time for.
We will help you generate vector embeddings***** and show you how to query your data******. Then you will be able to focus on building the rest of the app to impress us all.
****** if you already have vector embeddings for your data ready, that is cool too – we will show you how to load your data with your embeddings.* ******warning: you may be surprised how little code it takes to do it all.
🔨 Judgment Criteria
Overall delivery
Creativity
How well does it work?
Architecture
Does it scale? How close to prod is it? Enterprise readiness?
UX points
Intuitive? Is there a GUI? Compelling?
Weaviate Tool Use
💽 The Data:
Slack data from public MLOps Community channels
Transcripts of videos from MLOps.community Youtube page
Other data from the communities’ past events, blogs, etc. (make sure you share the sources used)
Starter Kit
🗺️ Location:
42 Berlin (exact location will be shared with participants on sign-up)
💫 Why we’re having this hackathon: As part of enhancing our community, we are looking to provide the opportunity to our members and other participants to learn to improve their knowledge on building LLM-powered solutions to a common problem of knowledge discovery by designing and building an app that uses LLMs to answer community member questions, just like the human experts.
Here’s everything you need to know to get started:
👷 You can work individually or in groups of up to 4 people
🔨 You can use any Open Source tools to build your solutions and your code need not be open-source.
💽 The data provided for the purposes of this hackathon is under the CC-BY-NC 4.0 license, intended for non-commercial use only.
📅 Projects should be started at the event kick-off to maintain fairness across projects
🔨 Get a head start with this vector database workshop hosted by Weaviate on Wednesday before the hackathon with this free workshop on Weaviate workshop on Wednesday
🔢 Credits: To be announced.
💸 Cash prizes: To be announced.
💫 Looking for creative solutions in the LLM App Stack (UI, System Design, Microservices) that solve real-world problems using emergent LLM technologies.
Agenda
To be announced.
Thank you to our sponsor Weaviate!
Disclaimers: Any of the details of this event are subject to change at the discretion of the organizer. The organizer of this event is not responsible for any damages, injuries, or legal consequences resulting from your attendance or behavior during the event. We reserve the right to update or modify this disclaimer at any time without prior notice. Your continued attendance at the event following any changes to the disclaimer constitutes your acceptance of those changes.
