Say Hi to Askie 🤖: Your Personalized AI Clone Chatbot

📚 Upload. 🌐 Link. ❓ Ask. 💡 Answer.
Meet Askie – your curious companion, not just a bot but your knowledge champion!

---

🎵 Who is Askie?

Have a doc or link to scan?
Askie's here to lend a hand.
PDF or a website page,
She’ll read it all, word by phrase.

---

What Askie Does

Askie is a Retrieval-Augmented Generation (RAG) based chatbot that allows users to:

• 1. Upload a PDF or enter a URL
• 2. Automatically chunk and vectorize the content using FAISS + HuggingFace Embeddings
• 3. Build a retrieval chain that fetches relevant chunks based on your queries
• 4. Generate natural, contextual answers using the Mistral model via Ollama
• 5. View the exact source documents used to answer your query
• 6. Optionally log interactions using Arize AI for observability

---

Tech Stack

Component	Tech Used
LLM	Mistral via Ollama (Local)
Embeddings	sentence-transformers/all-MiniLM-L6-v2
Vector Store	FAISS
Chunking	RecursiveCharacterTextSplitter
PDF Parsing	PyPDFLoader
Web Scraping	WebBaseLoader
Frontend	Streamlit
Observability	Arize AI (optional)

---

📁 Project Structure

Askie/
├── app.py              # Streamlit frontend
├── rag_pipeline.py     # RAG chain logic
├── utils/
│   └── chunking.py     # Chunking logic for PDF & URLs
├── .env                # Arize API keys and configs (should be excluded from GitHub)
├── data/               # Uploaded PDF files (usually excluded or kept empty)
└── README.md           # Project documentation

---

How Askie Works (Step-by-Step)

1️⃣ Upload or Input

• Upload a .pdf file OR enter a webpage URL.
• Askie reads and parses the content.

2️⃣ Chunk and Vectorize

• The content is chunked into small segments using RecursiveCharacterTextSplitter.
• Each chunk is converted into vector embeddings using HuggingFaceEmbeddings.

3️⃣ Build RAG Pipeline

• Chunks are stored in a FAISS vector store.
• A retriever fetches the most relevant chunks when a question is asked.
• Mistral LLM via Ollama generates an answer based on these chunks.

4️⃣ Chat and Source Tracking

• Askie responds to your question in natural language.
• Displays source document chunks used in the response.
• (Optional) Logs prompt, response, and metadata to Arize.

---

Local Setup

Prerequisites

• Ollama installed and running locally

Install Dependencies and Run!

pip install -r requirements.txt

Run Locally: streamlit run app.py

🙌 Made With Love By Aarya Shetiye