Interchat

Real-time AI collaboration — powered by GPT-4o, RAG, and seamless team workflows.

Live App · Features · Tech Stack · Local Setup · Environment Variables

Overview

Interchat is a full-stack, real-time AI collaboration platform that embeds AI assistants directly into team conversations. Users can chat with teammates, spin up AI-powered workspaces, and pull context from Google Drive, GitHub repositories, or their own conversation history — all in one place.

Live: https://interchat.app

Features

Real-Time Messaging

Direct 1-to-1 conversations between users
Messages delivered instantly via Socket.io
File attachments with Google Cloud Storage backing
Conversation history with efficient pagination

AI Chat (per Conversation)

Each conversation has its own AI chat session powered by OpenAI
Choose between GPT-4o-mini (fast) and GPT-4o (powerful)
Streaming responses with smooth word-level output
Auto-generated chat titles from the first message
Full message history persisted in MongoDB

Workspaces (Collaborative AI Spaces)

Shared AI workspaces for multi-user teams
Invite collaborators directly or by email (even before they sign up)
Switch AI models mid-conversation
Real-time updates for all participants via Socket.io

Context Manager (RAG-Powered)

Pull knowledge into every workspace from three sources:

Source	Description
Conversations	Semantically search past message threads
Google Drive	Index PDFs, Docs, Sheets, and Presentations
GitHub	Browse repositories and index source files

The AI uses these as retrieval-augmented generation (RAG) context when answering questions. Powered by Pinecone vector search with BGE Reranker v2-m3 for relevance ranking.

Friends & Social

Add friends by email address
Send and accept/reject friend requests
View friend profiles
Workspaces can be shared with anyone in your network

Authentication

Enterprise SSO via Auth0
Secure JWT sessions with NextAuth
Google Cloud file uploads scoped per user
GitHub OAuth for repository access (tokens stored in HTTP-only cookies)

Tech Stack

Layer	Technology
Framework	Next.js 15 (App Router), React 19, TypeScript 5
Styling	Tailwind CSS, Radix UI, Motion
Real-time	Socket.io 4 (dedicated server on port 4000)
Auth	Auth0 + NextAuth 4
Database	MongoDB Atlas
AI / LLM	OpenAI GPT-4o / GPT-4o-mini via Vercel AI SDK
Vector DB	Pinecone (RAG context)
Embeddings / Reranking	LlamaIndex, LangChain, BGE Reranker v2-m3
File Storage	Google Cloud Storage
Integrations	Google Drive API, GitHub REST API (Octokit)
Email	Resend
Deployment	Vercel

Project Structure

interchat_showcase/
├── app/
│   ├── (with-sidebar)/         # Protected routes (sidebar layout)
│   │   ├── conversations/      # DM threads
│   │   ├── workspaces/         # Collaborative AI workspaces
│   │   ├── friends/            # Friend management & profile
│   │   ├── profile/            # User settings
│   │   └── feedback/           # Feedback form
│   ├── api/
│   │   ├── auth/               # NextAuth handlers
│   │   ├── chat/               # LLM + workspace AI endpoints
│   │   ├── github/callback     # GitHub OAuth callback
│   │   └── upload/files        # Google Cloud file upload
│   └── components/             # 60+ shared UI components
├── lib/
│   └── tools/                  # RAG search tools (Pinecone)
├── types/                      # TypeScript definitions
├── public/images/              # Static assets & screenshots
├── scripts/                    # DB migration scripts
└── socketServer.ts             # Dedicated Socket.io server

Local Setup

Prerequisites

Node.js 18+
MongoDB Atlas account (free tier works)
Auth0 account
OpenAI API key
Pinecone account (for RAG features)
Google Cloud project (for file uploads and Drive integration)
GitHub OAuth App (for GitHub context)
Resend account (for emails)

1. Clone and install

git clone https://github.com/your-username/interchat_showcase.git
cd interchat_showcase
npm install

2. Configure environment variables

Copy the template below into a .env.local file and fill in your values:

cp .env.example .env.local   # or create .env.local manually

See the Environment Variables section for the full list.

3. Start the Socket.io server

Interchat uses a dedicated real-time server running on port 4000. Start it in one terminal:

npm run start:socket

4. Start the Next.js dev server

In a second terminal:

npm run dev

Open http://localhost:3000 in your browser.

5. Sign in

Navigate to http://localhost:3000 and sign in via Auth0. A user record is automatically created in MongoDB on first login.

Environment Variables

Create a .env.local file in the project root with the following variables:

# ── NextAuth ──────────────────────────────────────────────────────────────────
NEXTAUTH_SECRET=your_nextauth_secret
NEXTAUTH_URL=http://localhost:3000

# ── Auth0 ─────────────────────────────────────────────────────────────────────
AUTH0_SECRET=your_auth0_secret
AUTH0_BASE_URL=http://localhost:3000
AUTH0_ISSUER_BASE_URL=https://YOUR_DOMAIN.auth0.com
AUTH0_CLIENT_ID=your_auth0_client_id
AUTH0_CLIENT_SECRET=your_auth0_client_secret
NEXT_PUBLIC_AUTH0_CLIENT_ID=your_auth0_client_id
NEXT_PUBLIC_AUTH0_ISSUER_BASE_URL=https://YOUR_DOMAIN.auth0.com

# ── MongoDB ───────────────────────────────────────────────────────────────────
MONGODB_URI=mongodb+srv://<username>:<password>@<cluster>.mongodb.net/<dbname>

# ── OpenAI ────────────────────────────────────────────────────────────────────
OPENAI_API_KEY=sk-...

# ── Pinecone (RAG) ────────────────────────────────────────────────────────────
PINECONE_API_KEY=pcsk_...
PINECONE_INDEX_NAME=messages-index

# ── Google Cloud Storage ──────────────────────────────────────────────────────
GOOGLE_CLOUD_PROJECT_ID=your_project_id
GOOGLE_CLOUD_BUCKET_NAME=your_bucket_name
GOOGLE_CLOUD_KEY_JSON={"type":"service_account",...}   # service account JSON (stringified)

# ── Google Drive / OAuth ──────────────────────────────────────────────────────
GOOGLE_CLIENT_ID=your_google_client_id.apps.googleusercontent.com
GOOGLE_CLIENT_SECRET=GOCSPX-...
GOOGLE_API_KEY=AIzaSy...

# ── GitHub OAuth ──────────────────────────────────────────────────────────────
NEXT_PUBLIC_GITHUB_CLIENT_ID=your_github_oauth_app_client_id
GITHUB_CLIENT_SECRET=your_github_oauth_app_client_secret

# ── Email (Resend) ────────────────────────────────────────────────────────────
RESEND_API_KEY=re_...

# ── Socket.io ─────────────────────────────────────────────────────────────────
NEXT_PUBLIC_SOCKET_URL=http://localhost:4000
SOCKET_PORT=4000

Available Scripts

Command	Description
`npm run dev`	Start Next.js development server (port 3000)
`npm run start:socket`	Start Socket.io real-time server (port 4000)
`npm run build`	Build for production
`npm run start`	Start production server
`npm run lint`	Run ESLint

Both npm run dev and npm run start:socket must be running simultaneously for full functionality.

Database Schema (MongoDB)

Interchat uses a single MongoDB database (messaging_app) with the following collections:

Collection	Purpose
`users`	User profiles, friends list, pending invites
`conversations`	DM threads between participants
`messages`	Individual messages within conversations
`llm_chats`	AI chat sessions linked to conversations
`llm_messages`	Messages within AI chat sessions
`workspaces`	Collaborative AI workspaces
`workspace_messages`	Messages within workspaces

Architecture

Browser (React / Next.js)
        │
        ├── HTTP/REST ──► Next.js API Routes (/api/*)
        │                        │
        │                        ├── MongoDB Atlas (persistence)
        │                        ├── OpenAI API (LLM streaming)
        │                        ├── Pinecone (vector search / RAG)
        │                        ├── Google Cloud Storage (file uploads)
        │                        ├── Google Drive API (document import)
        │                        └── GitHub REST API (repo import)
        │
        └── WebSocket ──► Socket.io Server (port 4000)
                                 │
                                 └── Rooms: conversations · llm-chats · workspaces

Screenshots

Landing Page	Workspace View
AI Conversation	Friends & Profile

Name		Name	Last commit message	Last commit date
Latest commit History 183 Commits
app		app
hooks		hooks
lib		lib
public		public
scripts		scripts
types		types
.gitignore		.gitignore
README.md		README.md
components.json		components.json
eslint.config.mjs		eslint.config.mjs
middleware.ts		middleware.ts
next.config.ts		next.config.ts
package-lock.json		package-lock.json
package.json		package.json
postcss.config.mjs		postcss.config.mjs
socketServer.ts		socketServer.ts
tailwind.config.ts		tailwind.config.ts
tsconfig.json		tsconfig.json
tsconfig.server.json		tsconfig.server.json

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Repository files navigation

Interchat

Overview

Features

Real-Time Messaging

AI Chat (per Conversation)

Workspaces (Collaborative AI Spaces)

Context Manager (RAG-Powered)

Friends & Social

Authentication

Tech Stack

Project Structure

Local Setup

Prerequisites

1. Clone and install

2. Configure environment variables

3. Start the Socket.io server

4. Start the Next.js dev server

5. Sign in

Environment Variables

Available Scripts

Database Schema (MongoDB)

Architecture

Screenshots

About

Uh oh!

Releases

Packages

Uh oh!

Contributors

Uh oh!

Languages

Folders and files

Latest commit

History

Repository files navigation

Interchat

Overview

Features

Real-Time Messaging

AI Chat (per Conversation)

Workspaces (Collaborative AI Spaces)

Context Manager (RAG-Powered)

Friends & Social

Authentication

Tech Stack

Project Structure

Local Setup

Prerequisites

1. Clone and install

2. Configure environment variables

3. Start the Socket.io server

4. Start the Next.js dev server

5. Sign in

Environment Variables

Available Scripts

Database Schema (MongoDB)

Architecture

Screenshots

About

Resources

Uh oh!

Stars

Watchers

Forks

Releases

Packages 0

Uh oh!

Contributors

Uh oh!

Languages

Packages