HealthHive is a real-time Q&A assistant built using Pathway’s streaming platform and RAG (Retrieval-Augmented Generation) techniques. It continuously ingests live data (support tickets, chats, documents), indexes it in real-time, and provides context-aware answers to user queries using LLMs. This assistant ensures enterprise teams receive answers grounded in the most current data.
We are Developing a real-time assistant that ingests enterprise data streams and offers Retrieval-Augmented Generation (RAG) based answers using the latest information.
This project solves the need for:
- Instant insights from incoming data (tickets, docs, chats).
- Reliable, up-to-date answers.
- Seamless integration with enterprise workflows.
- 🔎 Employees ask technical or support questions via the web UI or Slack.
- 📥 Data (tickets, documentation updates, FAQs) is streamed and indexed in real-time using Pathway.
- 🧠 RAG pipeline + LLM retrieves and answers based on the most relevant, recent context.
- 📡 Results are streamed back instantly to the user via the interface or API.
+-------------------+ +------------------+ +-----------------------+
| Data Sources | ---> | Pathway Pipeline | ---> | Vector Store (Chroma) |
| (Docs, Tickets) | | | | |
+-------------------+ +------------------+ +----------+------------+
|
+----------------v--------------+
| LLM + Retrieval (RAG Pipeline)|
+----------------+--------------+
|
+------------v-----------+
| API / UI (React) |
+------------------------+