Ai4ALL Hackathon Wrap Up
After an exciting couple of weeks that saw the submission of 44 different projects from a total of 173 participants, the AI4All Hackathon has wrapped and the winners have been selected.
We had participants from all over the world including UAE, India, Saint Lucia, Guatemala, South Africa, USA, Germany, Nigeria, Serbia, Poland, Canada, Italy, France, Georgia, Brazil, Kenya, Argentina, Belgium, Czech Republic, Spain, South Korea, UK, Colombia, Ireland, Scotland, Pakistan, Finland, and Ghana.
Respected personalities from both the Bitcoin and AI industries served as judges,and after careful consideration, they awarded a total of more than $70,000 in cash and other prizes to participants whose projects focused on one of four tracks: Agents (Machine Payments) Track, Privacy Track, Training Track, and Bitcoin Education Track.
Overall Winner: PhotoBolt
PhotoBolt converts a basic product image into a marketing poster by leveraging Nostr Data Vending Machines to break down the work into a chain of smaller jobs that AI agents compete against one another to service. WebLN, an open standard for bitcoin payments on the web, is used to seamlessly pay these AI agents for doing their jobs.
Agents Track Winner: Flow
Flow enables less-technical people to use the n8n platform to build no-code workflows that leverage AI-based automation. People without programming skills or bitcoin knowledge can now monetize these workflows with a combination of Nostr Wallet Connect and L402, which is abstracted away via Alby Wallet.
Privacy Track Winner: Netonomy
Netonomy is a web portal that attempts to allow users to control their own digital identity and data and selectively reveal it to various applications. It includes a personal data store built from a combination of nostr NIP-07, decentralized identifiers (DIDs), and decentralized web nodes (DWNs). The app also allows users to post jobs for AI agents to complete via Nostr Data Vending Machines, and it integrates an OpenAI chatbot powered by an L402 wrapper.
Training Track Winner: Bitcoin-PAL
Bitcoin-PAL is a bitcoin-focused AI chatbot coupled with an incentivized crowd-sourcing platform to train and educate LLMs and users with bitcoin payouts. Contributors are paid bitcoin for helping to train the model on high-quality documentation/resources, and chatbot users receive responses based on vetted, bitcoin-focused information.
Bitcoin Education Track Winner: BitDevs Upgraded
BitDevs Upgraded is a new website template with an improved UI and auto-generated summaries of meetup content. This makes it easier for new BitDevs meetups to spin up useful websites out-of-the-box, and attendees can get an idea of the content without reading the full article.
Other Projects
"These might be the winners, but the creativity of projects was mind blowing. Have a look at these live, up and running applications," Alby wrote on its hackathon recap blog post.
- Transcribe.fm: "A transcription service for podcasters with speaker diarization & word-by-word fidelity where you can pay-what-you-want with your Alby Lightning wallet. Also available as an L402 API."
- PlebAI: "A free and open Source ChatGPT with open source Large Language Models payable with bitcoin via WebLN."
- Matador: "A service that lets you sell API access against an arbitrary API using your API Key in exchange for bitcoin micropayments via L402 Payment Required Codes."
- SmartSats: Mobile lightning wallet that allows you to conveniently allocate sats for AI agents, enabling task execution within predefined budgets, with the agent having the ability to withdraw funds as needed.
- Nostr Serverless API: "Typically, data scientists access social media data through well-established APIs and their associated wrappers. We've introduced the Nostr Serverless API (NSA) to recreate this familiar user experience for Nostr."
- Bitcoin TLDR: "Summaries of posts and threads in the bitcoin-dev and lightning-dev mailing lists."
- NiNa Ai: Lightning node management insights and automation tool.
- Check out all 44 hackathon projects and their pitches here.