Multi-AgentAI
MeetingCopilot

An autonomous gRPC-based agent ecosystem that joins your meetings, understands context, and manages your workflow.

0
AI Agents
Specialized & isolated
0%
A2A Native
gRPC message bus
0+
Tech Integrations
End-to-end pipeline
0
Core Platforms
Web · Extension · API

The Architects

Built by a specialized two-person team.

D

Darshan Gowda G S

1st AuthorLead Author

Led the full-stack architecture — Next.js Material 3 dashboard, FastAPI backend, A2A gRPC communication layer, Google ADK agent integration, and the Chrome Extension.

Domains

Next.js / React
FastAPI
Chrome Extension
A2A / gRPC
Material 3 UI

Tech Stack

Next.js 14
React
TypeScript
FastAPI
Python
gRPC
Supabase
Redis
K

Kanika Pitaliya

2nd AuthorCo-Author

Designed the multi-agent AI ecosystem — semantic memory architecture with Qdrant, Lyzr orchestration, LLM agent prompting strategies, and AI pipeline research.

Domains

Google ADK
Qdrant
Lyzr
LLM Research
AI Pipelines

Tech Stack

Google ADK
Lyzr
Qdrant
Vector Embeddings
LLM Prompting
A2A Protocol
OpenTelemetry
Langfuse

Core Features

Everything you need to automate, remember, and manage your meetings without leaving the context.

Multi-Agent Intelligence

Specialized AI agents collaborate to perform dedicated tasks instead of relying on a single monolithic LLM.

Real-Time Assistance

Contextual suggestions, meeting insights, and intelligent support while the meeting is still in progress.

Semantic Memory

Store and retrieve meeting knowledge using vector embeddings powered by Qdrant.

Smart Meeting Summaries

Automatically generate concise summaries, key discussion points, and actionable takeaways.

AI Follow-ups

Generate professional follow-up emails containing meeting highlights and action items.

Intelligent Scheduling

Create reminders and follow-up meetings directly from extracted action items.

Document Question Answering

Upload supporting documents and let AI agents answer questions using meeting context.

Late Join Recaps

Users joining late receive an instant AI-generated summary of everything discussed so far.

Agent Ecosystem

Our A2A (Agent-to-Agent) architecture uses gRPC for high-speed, isolated communication between specialized AI modules.

Transcription

Streams and processes live meeting transcripts.

Realtime

Generates contextual suggestions mid-meeting.

Summary

Produces summaries, key points, action items.

Memory

Stores semantic embeddings in Qdrant.

Email

Drafts follow-up emails from meeting context.

Scheduler

Converts action items to calendar events.

Docs QA

Answers questions using uploaded documents.

Late Join

Instant recap for mid-meeting participants.

Orchestrator

Routes and coordinates all agents via A2A.

MeetMaxxing
2024 MeetMaxxing. Open Source Project.