hackquest logo

TechTrio

decentralized AI-powered shopping assistant combining top tech influencers' expertise with up-to-date tech insights

视频

描述

DEMO: https://youtu.be/NSBnXdprg5k


## Overview

TechTrio Gaia Multiagent is a cutting-edge Living Knowledge System that leverages the power of decentralized AI to provide users with comprehensive, up-to-date insights on the latest tech products. By combining the expertise of top tech influencers, real-time product information, and advanced AI models, TechTrio offers an unparalleled shopping assistant experience for tech enthusiasts.

<img width="889" alt="excalidraw" src="https://github.com/user-attachments/assets/0ec3de4b-7db9-438f-8c90-050d023011bf">

### Key Features

* **Multi-Agent System**: Utilizes AI agents modeled after three renowned tech reviewers: MKBHD, Unbox Therapy, and MrWhosTheBoss.

* **Living Knowledge Base**: Employs Upstash Vector DB as an inventory system to continuously update information on gadgets and new devices.

* **Decentralized Infrastructure**: Built on GaiaNet nodes, ensuring a robust and scalable system.

* **Impartial Moderation**: Incorporates a moderator agent to summarize and compare reviews, providing balanced conclusions.

### Technical Stack

* **AI Models**: Utilizes various models including Llama, Phi, and Gemma through GaiaNet nodes.

* **Vector Database**: Employs Upstash Vector DB for efficient storage and retrieval of product information.

* **Embedding**: Leverages the bge-base-en-v1.5 model for text embedding.

* **API Integration**: Seamlessly integrates with OpenAI-compatible APIs provided by GaiaNet.

### How It Works

1. **User Query**: The user submits a question about a tech product they're interested in.

2. **Multi-Agent Processing**: The query is processed by AI agents representing MKBHD, Unbox Therapy, and MrWhosTheBoss, each providing their unique perspective.

3. **Knowledge Retrieval**: The system queries the Upstash Vector DB to retrieve the most up-to-date product information.

4. **Review Generation**: Each AI agent generates a review based on their persona and the available information.

5. **Moderation and Summary**: A moderator agent compiles and summarizes the reviews, highlighting key points of agreement and disagreement.

6. **Final Output**: The user receives a comprehensive summary of the product reviews, along with a balanced conclusion to aid in their purchasing decision.

### Key Advantages

* **Diverse Perspectives**: Gain insights from multiple trusted tech influencers in one place.

* **Up-to-Date Information**: The living knowledge system ensures that product information is always current.

* **Balanced View**: The moderator summary provides an impartial overview of different opinions.

* **Time-Saving**: Quickly access expert opinions without watching multiple video reviews.

* **Personalized Recommendations**: Receive tailored advice based on user preferences and needs.

### Future Development Roadmap

* **Expanded Expertise**: Integrate with more tech influencers and experts to broaden the knowledge base.

* **Product Category Expansion**: Cover a wider range of tech products and categories to cater to diverse user needs.

* **User Feedback Mechanisms**: Implement user feedback systems to improve AI responses and enhance the overall user experience.

* **Price Comparison and Availability Features**: Add features to compare prices and check availability of products to facilitate informed purchasing decisions.

本次黑客松进展

research gaianet, research technical docs. understand possibilities. go through ideas. brainstrom ideas, define the idea. look for tools to build the idea. try and error with features implementations. readjust ideas as progress in the development

队长Aaeither
赛道
AIOther