ThirdSpace is an AI-powered networking tool that transforms spontaneous encounters in "third spaces" (hackathons, conferences, meetups, cafes, etc.) into meaningful and trackable relationship opportunities.

Instead of forgetting conversations after events, ThirdSpace helps users capture the interaction and receive AI-generated insights on how to build stronger connections.

Networking in real life is often spontaneous and unstructured. People meet briefly at events, exchange ideas, and then lose touch because there is no system to capture the context of the interaction.

Important information like:

• what you talked about
• shared interests
• why the connection matters
• how to follow up

is often forgotten.

ThirdSpace provides a simple workflow:

1. A user scans a QR code in a shared environment (a "third space").
2. They create an account and log their interaction with another person.
3. The interaction details are sent to an AI backend.
4. The AI analyzes the relationship and generates actionable insights.

The system returns:

• a connection summary
• shared interests
• connection strength
• a suggested follow-up message
• recommended next steps

This turns casual encounters into intentional networking opportunities.

The AI component is powered by an OpenAI-based backend built with Node.js.

When a user submits a connection, the frontend sends the following information to the backend API:

• Person A profile (user)
• Person B profile
• third-space context
• event or location
• conversation notes
• networking goal

Example input sent to the AI: Person A: Renée, software engineering intern interested in AI Person B: Shriya, startup founder Third space: Hackathon Event: Lunaris Hacks Conversation notes: Discussed AI startups and product design Goal: Friendship

This data is sent to the /analyze endpoint of the backend server.

The backend constructs a structured prompt and sends it to the OpenAI API using the gpt-4o-mini model.

The AI is instructed to return structured JSON with the following format: { "summary": "2-4 sentence connection summary", "sharedThirdSpace": "shared context description", "sharedInterests": ["interest1", "interest2"], "strength": "Weak / Moderate / Strong", "followUp": "Suggested follow-up message", "recommendedAction": "Recommended next step" }

The backend then parses the AI response and sends it back to the frontend.

The frontend is built with:

• HTML
• CSS
• JavaScript

Features include:

• user signup and login
• connection logging form
• AI-generated relationship insights dashboard
• shared interest tags
• suggested follow-up messages

The UI uses a dark space-inspired theme to reflect the idea of navigating connections in the "Third Space".

Frontend

• HTML
• CSS
• Vanilla JavaScript

Backend

• Node.js
• Express

AI

• OpenAI API (gpt-4o-mini)

Storage

• Browser LocalStorage (prototype)

project │ ├── frontend │ ├── index.html │ ├── connect.html │ ├── login.html │ ├── signup.html │ ├── result.html │ ├── app.js │ └── style.css │ └── server └── server.js

1. User logs into ThirdSpace.
2. They record a connection made in a shared third space.
3. The frontend sends the data to the backend API.
4. The backend generates a prompt and sends it to the OpenAI model.
5. The AI analyzes the interaction.
6. Structured insights are returned to the frontend.
7. The user receives:

• connection summary
• shared interests
• connection strength
• suggested follow-up message
• recommended next step

• QR code scanning integration
• persistent database (MongoDB or Firebase)
• LinkedIn integration for networking
• calendar-based follow-up reminders
• event-based networking analytics

Built by the HackHerz team.