Inspiration
Originally, PAST was a tool that allowed you to send money to anyone anywhere, without knowing who they were! But we realized there is so much more potential.
There is so much data about you on the internet. From the blogs you write to the images you post. What if, instead of using your data maliciously, we could use it to create more personalized connections with people you meet every day?! PAST is a tool that searches the web for your public social media, LinkedIn, contact information, and other public data to create personalized connections, workflows, and people search.
What it does
PAST uses a series of agents and workflows to do background research on who you are from a scan of your face. We then display all of this information in a central HUD where you can find out new information about the person you are meeting! After the fact, you can use PAST's agents to search for the people you have met, bridging the gap of cordiality.
How we built it
We used a plethora of tools across a variety of frameworks and industries.
- OpenCV was used to handle the vision aspect of the product; the backbone of the project
- The SUI and XRP L1s were used for the transfer of money
- The Lava RPC was used for L1 integrations across the EVM stack
- Poke (Interaction) was used as the messaging foundry for our people MCP
Challenges we ran into
A lot -- talk to us if you want to get the full story D:
Accomplishments that we're proud of
PAST is fully functional and deployed across 3 repositories, as well as 5 different microservices! This is a fully functional product!

Log in or sign up for Devpost to join the conversation.