Back
Explore EventsExplore ProjectsMy Projects
BotForge — screenshot 1
BotForge — screenshot 2

BotForge

Transform Topics and Documents into Retrieval-Powered AI Experts.

Codex Community Hackathon - Pune

Links

Repository

github.com/Yug-11/BotForge

Demo video

youtu.be/9g9KpQ1IKCI?si=COXOyWlRD0zKdFjx

Team

2 members
  • SU

    Sujal Tamboli

    Owner
  • YU

    Yug Ketkale

Overview

BotForge AI

Tagline

Transform Topics and Documents into Retrieval-Powered AI Experts. Description

BotForge AI is a RAG-powered platform that enables users to create custom AI experts from topics, web knowledge, and documents without any technical expertise.

Traditional AI chatbots rely solely on pre-trained knowledge, which can lead to outdated information and hallucinated responses. BotForge AI addresses this challenge by combining Retrieval-Augmented Generation (RAG), vector search, and Large Language Models to provide accurate, context-aware answers grounded in real knowledge sources.

The platform supports multiple knowledge acquisition methods:

Topic Bots– Users can provide a topic, and the system automatically researches the web, collects relevant information, generates embeddings, and builds a specialized AI expert. PDF Bots– Users can upload documents such as syllabi, research papers, reports, or manuals, allowing the system to extract, process, and understand document-specific knowledge. Hybrid Knowledge Bots – Users can combine web research with uploaded PDFs to create AI assistants that leverage both public and private knowledge sources.

How It Works

  1. The system gathers knowledge from user-provided topics and uploaded documents.
  2. Content is cleaned, chunked, and converted into vector embeddings.
  3. Embeddings are stored in a Qdrant vector database.
  4. When a user asks a question, the system retrieves the most relevant information using semantic search.
  5. Retrieved context is provided to the LLM, enabling grounded and accurate responses.

Key Features

  • RAG-based intelligent question answering
  • Topic-driven AI expert creation
  • PDF document understanding and chat
  • Hybrid knowledge integration (Web + PDF)
  • Semantic search with vector embeddings
  • Persistent chat history per bot
  • Knowledge refresh and re-training capability
  • Multi-bot management dashboard
  • Reduced hallucinations through retrieval-based context injection

Technology Stack

Frontend

  • React
  • Tailwind CSS
  • Axios

Backend

  • FastAPI
  • Python

AI & RAG

  • OpenAI GPT Models
  • OpenAI Embeddings
  • Qdrant Vector Database
  • Retrieval-Augmented Generation (RAG)

Impact

BotForge AI empowers students, researchers, professionals, and organizations to transform information into intelligent conversational assistants. Instead of manually searching through large documents or scattered online resources, users can create domain-specific AI experts capable of delivering fast, reliable, and context-aware answers.

By combining web intelligence, document understanding, and modern RAG architecture, BotForge AI makes advanced AI knowledge systems accessible to everyone.

ExploreProjectsMine