Hackathon Portal
AI Tinkerers - San Francisco
Sign in

Agents with Superpowers: Context Engineering Hackathon

Sign in
Founding Engineer - Stealth
Build healthcare AI agents with Claude Code. Apply for remote roles.
DevRel Lead - Restate
Build durable AI agent infrastructure. Drive US remote adoption today.
All Jobs
Final round winners have been announced. View Results
Team

sambanova

Project Concept

The Sambanova Voice AI Productivity System is an enterprise-grade platform that combines LangGraph AI agents, team collaboration, voice interaction, and intelligent call center integration. Built for hackathon demonstrations, it showcases advanced features including Redis audio stream and composio integration.

Entry

Status: Submitted

Last saved: October 25 at 5:51 PM PDT

View Entry

Team Roster

Message board not available for this team yet.

Hyeokjin Lee Team Lead RSVP Approved

software engineer at hjlees
ools Used Core Technologies Flask - Web framework for the main application Socket.IO - Real-time communication for WebRTC voice assistant Redis - Session management, caching, Pub/Sub for real-time notifications Composio - External tool integrations (Slack, GitHub, Gmail, Notion, Jira) LangGraph - AI agent orchestration OpenAI APIs - Whisper for transcription, TTS for speech generation WebRTC - Real-time voice communication SQLAlchemy/PostgreSQL - Database for todos, reminders, calendar events Development & Deployment Tools Render.com - Cloud deployment platform Git - Version control MCP (Multi-tool Co-Processor) - Framework for agent tools Gunicorn - WSGI server for production Specific Components Built 1. WebRTC Voice Assistant System File: sambanova/webrtc_voice_server.py Features: Real-time audio streaming and processing Audio format detection (WebM, WAV, MP3, etc.) Base64 encoding/decoding for Redis storage WAV fallback mechanism for raw PCM audio Integration with AI agent for voice commands 2. Redis Integration & Management File: sambanova/redis_manager.py Features: Session management Audio buffer storage Pub/Sub for real-time notifications Rate limiting and analytics Robust environment variable parsing 3. Composio External Tool Integration File: sambanova/composio_tools.py Features: Slack message sending GitHub, Gmail, Notion, Jira integrations Robust method discovery for API compatibility Fallback mechanisms for tool loading 4. Audio Stream Players Files: audio_stream_player.py - Full-featured player simple_audio_player.py - Simplified version sambanova/audio_player_routes.py - Flask Blueprint integration Features: Real-time audio playback from Redis Multiple audio format support Download functionality Web interface for audio analysis 5. AI Agent System File: sambanova/assistant_graph_todo.py Features: TodoAgent class with LangGraph orchestration Tool calling capabilities Voice command processing Integration with MCP tools 6. MCP Tool Integration File: sambanova/mcps/local_servers/db_todo.py Features: Todo creation with Slack notifications Team todo management Google Calendar integration Database operations with error handling 7. Voice Command Processing File: sambanova/routes.py Features: WebRTC voice processing Twilio voice integration Agent timeout handling Audio transcription and TTS generation 8. Frontend Integration File: templates/webrtc_voice_assistant.html Features: WebRTC client interface Audio recording and playback Multiple audio format support Real-time communication with backend
experienced software engineer with 25+ years of contact center industry
Hybrid voice AI contact center
https://hjlees.com/sambanova-tech-spec