🚀 Inspiration

In large tech communities like AI collectives, Discord, and Slack groups, finding the right collaborator, co-founder, or even a relevant opportunity often turns into chaos. People introduce themselves in long-forgotten #introduction threads, and most great matches never happen simply because no one has the time or patience to read through them. We realized this wasn’t just a one-time issue—it’s a broken system. That’s where the idea for Concisa.ai was born: a conversational, AI-powered matchmaking platform that understands your intent and connects you to the right people, jobs, events, or companies—without the noise.


💡 What it does

Concisa.ai is an AI-powered matchmaking platform with a chat-based interface where professionals create rich profiles during onboarding. Users can then interact with a conversational assistant that helps them find:

  • The top 3 people aligned with their collaboration needs
  • Jobs tailored to their skills and preferences
  • Events relevant to their domain or goals
  • Companies that match their values or career aspirations

No more endlessly scrolling—Concisa.ai intelligently curates results based on deep profile understanding and user intent.


🛠️ How we built it

  • Frontend: Built using React.js with Tailwind CSS for clean, responsive UI components
  • Backend: Python + FastAPI for managing user profiles, matchmaking logic, and integrations
  • Vector Store: Qdrant for storing and retrieving embeddings of users, jobs, and events
  • Embedding Model: OpenAI’s text-embedding-3-small for matching intent with profiles and listings

* Chat Layer: AI assistant powered by OpenAI for natural language conversations

🧩 Challenges we ran into

  • Structuring embeddings in a way that captures contextual compatibility between people, not just skills
  • Handling profile sparsity when some users only input partial data
  • Designing a conversational flow that feels natural but is also efficient
  • Ensuring real-time search performance in Qdrant with dynamic filters across multiple entity types

🏆 Accomplishments that we're proud of

  • Successfully created a working demo that lets users find highly accurate matches with just a simple message
  • Unified people, jobs, companies, and events into a single AI-powered discovery pipeline
  • Made the process of team and opportunity discovery fun, fast, and frictionless

📚 What we learned

  • People don’t just need recommendations—they need guided, conversational discovery
  • Embeddings are incredibly powerful when combined with structured metadata
  • Matchmaking is as much about intent interpretation as it is about data similarity
  • Chat-based discovery lowers the barrier for action and makes search more human

🔮 What's next for Concisa.ai

  • WhatsApp & Discord Integration: We’re integrating Concisa.ai into existing communication platforms like WhatsApp and Discord so users can find matches right where they chat.
  • Dedicated Matchmaking Agent: A specialized agent where users can simply enter how many teammates they need—and it will return a fully compatible, AI-curated team.
  • Voice Interface: Adding a voice-based assistant for hands-free access to matchmaking
  • Analytics Dashboard: For teams and companies to visualize talent pools, event trends, and matchmaking metrics

Built With

  • agents
  • lovable
  • n8n
  • openai
  • supabase
Share this project:

Updates