Series Live

Voice AI-powered Omegle for iMessages. Series Live connects users through group calls, introduces participants with AI-generated intros, and keeps conversations flowing by participating during awkward pauses.

What It Does

• Smart Matchmaking: Text your interests and get matched with compatible people
• AI Introductions: When a call starts, AI introduces each participant based on their profiles
• Conversation Flow: AI participates during awkward silences to keep discussions engaging
• iMessage Integration: Everything happens through iMessage - no app download required

How It Works

1. Join the Waitroom: Send "Call" to enter the matchmaking queue
2. Get Matched: Based on interests, you'll be matched with others
3. Confirm: Review the match profile and reply "Yes" to connect
4. Get Called: Twilio calls both participants into a conference
5. AI Hosts: Voice AI introduces participants and keeps conversation flowing

Quick Start

1. Install Dependencies

pip install -r requirements.txt

2. Configure Environment

Create a .env file with:

# Series API (iMessage)
SERIES_API_KEY=<your_api_key>
SERIES_BASE_URL=<api_base_url>
SERIES_SENDER_NUMBER=<sender_number>

# Twilio (Voice Calls)
TWILIO_ACCOUNT_SID=<your_account_sid>
TWILIO_AUTH_TOKEN=<your_auth_token>
TWILIO_PHONE_NUMBER=<your_twilio_number>

# OpenAI (AI Features)
OPENAI_API_KEY=<your_openai_key>

# User Profiles (for demo)
USER1_PHONE=<phone1>
USER1_NAME=<name1>
USER2_PHONE=<phone2>
USER2_NAME=<name2>

# Webhook (for ngrok)
NGROK_URL=<your_ngrok_url>

3. Run the Server

python main.py

4. Start a Conversation

Send any of these commands via iMessage:

• Call - Enter the waitroom to find a match
• Leave - Exit the waitroom
• Status - Check your waitroom status
• List - See who's waiting
• Summary - Get AI summary of your last call
• Help - Show available commands

Project Structure

series/
├── main.py                   # Entry point
├── config.py                 # Environment loader
├── consumers/
│   └── kafka_consumer.py     # Kafka consumer for iMessage events
├── services/
│   ├── series_api.py         # iMessage API client
│   ├── twilio_service.py     # Conference call service
│   ├── matchmaking.py        # Interest-based matching
│   └── notetaker.py          # Call transcription & summary
├── handlers/
│   └── message_handler.py    # Command & event routing
└── requirements.txt          # Dependencies

Architecture

iMessage → Kafka → Message Handler → Matchmaking
                                   → Twilio Calls
                                   → AI Voice (Intro/Icebreakers)

Commands Reference

Command	Description
Call	Join the matchmaking waitroom
Leave	Exit the waitroom
Status	Check if you're in the waitroom
List	See who else is waiting
Summary	Get AI summary of your last call
Help	Show available commands
Yes	Accept a match
No	Decline a match

License

MIT