Endee Assignment Submission – RAG AI Chatbot

.gitignore

@@ -1,32 +1,6 @@
-
-# Ignore build directory
-build/
-# Ignore build like directory
-build*/
-# Ignore all files in the build directory
-build/*
-
-# Ignore tests build directory
-tests/build/
-tests/build*/
-
-# Test binaries
-tests/**/ndd_filter_test
-
-# macOS debug symbols
-*.dSYM/
-
-# Sometimes data files are created for tetsing
-data/
-data/*
-
-# VS Code directory
-.vscode/
-.vscode/*
-
-# Frontend
-frontend/
-frontend/*
-
-# DS Store
-.DS_Store
+.env
+backend/.env
+__pycache__/
+*.pyc
+node_modules/
+venv/

Knowledge_base.py

@@ -0,0 +1,97 @@
+SAMPLE_DOCS = [
+
+# 🔹 RAG
+{
+"title": "RAG",
+"content": """
+Retrieval-Augmented Generation (RAG) is an AI architecture that combines retrieval and generation.
+Instead of answering directly, the system first retrieves relevant information from a knowledge base.
+Then it uses a language model to generate a response based on that retrieved data.
+RAG improves accuracy, reduces hallucination, and makes responses more reliable.
+"""
+},
+
+# 🔹 VECTOR SEARCH
+{
+"title": "Vector Search",
+"content": """
+Vector search converts text into numerical embeddings.
+These embeddings represent the meaning of the text.
+Instead of matching keywords, vector search finds similar meanings.
+It is used in semantic search, recommendation systems, and chatbots.
+Vector search helps retrieve relevant context quickly and accurately.
+"""
+},
+
+# 🔹 ENDEE
+{
+"title": "Endee",
+"content": """
+Endee is a vector database used in this project.
+
+Purpose of Endee:
+- Store embeddings of text data
+- Perform similarity search
+- Retrieve relevant information for user queries
+
+In this project:
+1. User query is converted into embeddings
+2. Endee searches similar embeddings
+3. Relevant chunks are retrieved
+4. These chunks are passed to the AI model
+5. The AI generates answers based on retrieved context
+
+This enables Retrieval-Augmented Generation (RAG).
+"""
+},
+
+# 🔹 PROJECT ARCHITECTURE
+{
+"title": "Project Architecture",
+"content": """
+This project follows a RAG architecture.
+Step 1: User sends a query.
+Step 2: Query is converted into embeddings.
+Step 3: Vector database (Endee concept) retrieves relevant chunks.
+Step 4: Retrieved context is sent to the LLM.
+Step 5: LLM generates an answer based only on that context.
+"""
+},
+
+# 🔹 CHATBOT SPEED
+{
+"title": "Performance Optimization",
+"content": """
+Chatbot response speed can be improved by:
+1. Using efficient vector search
+2. Reducing API latency
+3. Using caching for repeated queries
+4. Optimizing backend processing
+5. Using faster LLM models
+6. Minimizing unnecessary computations
+"""
+},
+
+# 🔹 SEMANTIC SEARCH
+{
+"title": "Semantic Search",
+"content": """
+Semantic search focuses on understanding meaning instead of exact keywords.
+It uses embeddings to capture context.
+This allows the system to return relevant answers even if the words are different.
+It improves chatbot accuracy and user experience.
+"""
+},
+
+# 🔹 LIMITATIONS
+{
+"title": "Limitations",
+"content": """
+RAG systems depend on the quality of the knowledge base.
+If the information is not present, the system cannot answer.
+They also require proper chunking and embedding strategies.
+Performance depends on vector search efficiency.
+"""
+}
+
+]