Get together with the Opper community and build with LLMs, free access to tokens from a lot of different models (Model list), RAG, few shot prompting and other techniques to build with quality. Learn and share with friends.
A hackathon with Opper is a great way to experience building LLMs. As many of you have undoubtly have heard, it's quite easy to make a demo with LLMs but trying to build something that can become geared towards production grade, is quite tricky indeed. Opper is a platform that integrates the necessary tools in a single toolbox that enables you to build for production.
Opper (opper.ai) is an integrated platform to build production grade generative AI features, chatbots, and agents.
-
Build structured, model independent LLM calls. Interact with frontier, open-source, and multimodal models using a unified, structured syntax. We empower you to instruct models as code, and reduce the need for complex prompts.
-
Integrate your own data and knowledge. Opper enables both indexing and retrieval of the semantically relevant parts of data and connect it to calls, while you retain full integration control.
-
Trace every call and session for full visibility. Track every AI interaction: view sessions, calls, outputs, feedback, reasoning, and costs as easy to understand real-time traces.
-
Optimize quality, security, cost and speed in production. Automate LLM tuning with Opper Functions. Optimal models and configs, identified through continuous evaluations and feedback.
-
Build from a secure, compliant and European foundation. Opper is built for compliance: GDPR, SOC2, DORA, and EU-ready. Focus on value while we handle regulations.
Join our community on Discord Opper Community.
This is an on site hackathon. We offer online hackathons as well, be on the lookout for these.
If you are invited to this hackathon, please RSVP so that we can plan accordingly.
We tend to keep our hackathons to 4 hours. Generally, this boils down to:
- 30 mins intro and to get started in teams or individually
- Build for 3 hours
- 30 mins for show n tell
To be efficient with the time we have, we've made a checklist here.
Checklist for attendees:
- Create your Opper account here and generate the sample call when logging in.
- Decide how you want to integrate (Python SDK (pip install opperai), TypeScript SDK (npm install opperai), API).
- Know what to build. We have a list of examples below which includes techniques like RAG, few-shot prompting etc, to get you started.
- Join the Discord. Opper engineers are available to assist you in person, but the discord can be used both before, during and after the hackathon if you have any questions.
In case you don’t want to build from the ideas or examples below, feel free to create your own hack. But please, have this prepared in beforehand so that we don't spend the hack time to try to come up with an idea.
Show and tell together, let’s share learning and challenges. We usually bring some swag to hand out.
This list is in no way exhaustive, but these are some ideas on what you can build with generative AI. Just to serve as potential inspiration:
- Study Companion: Take a photo of a page and generate flash cards, optionally with voice input to test the user.
- Application Generator: Iteratively add instructions and have AI build an app for you
- Touchless Browsing: A chrome plugin that lets you drive the browser using voice input.
- Supercharge Google Sheets: Build a macro for Google Sheets to call AI models to fill in columns with AI.
- Visual Recipe Assistant: Take a photo of ingredients, get recipe suggestions, and receive audio instructions while cooking.
- Accessible Museum Guide: Point camera at exhibits to generate alt text for visually impaired users.
- Visual Math Solver: Take a photo of a math problem and receive step-by-step text explanations of the solution, with an ability to issue voice commands.
- Plant Identification and Care Assistant: Snap a photo of a plant to get care instructions as text.
We’ve compiled some example projects to get you started. Feel free to pick any one of them and go ahead. Please read below to get some info on focus areas.
| Focus area | Repo |
| Database integration, SQL natural lang assistant | https://github.com/opper-ai/example-sql-assistant |
| Simple chat portal built with Opper and Streamlit | https://github.com/opper-ai/example-streamlit-opper |