Add initial implementation of chat application with environment variable support and vector storage integration

.gitignore

@@ -1,5 +1,6 @@
 data/
 db/
+.chainlit/
 
 # Byte-compiled / optimized / DLL files
 __pycache__/

README.md

@@ -0,0 +1,22 @@
+# Welcome to TheDataGuy Chat! 👋
+
+This is a Q&A chatbot powered by TheDataGuy blog posts. Ask questions about topics covered in the blog, such as:
+
+- RAGAS and RAG evaluation
+- Building research agents
+- Metric-driven development
+- Data science best practices
+
+## How it works
+
+Under the hood, this application uses:
+
+1. **Snowflake Arctic Embeddings**: To convert text into vector representations
+2. **Qdrant Vector Database**: To store and search for similar content
+3. **GPT-4o-mini**: To generate helpful responses based on retrieved content
+4. **LangChain**: For building the RAG workflow
+5. **Chainlit**: For the chat interface
+
+## Sources
+
+All answers are generated based on content from [TheDataGuy blog](https://thedataguy.pro/blog/). Sources are shown for each response so you can read more about the topic.

app.py

@@ -0,0 +1,110 @@
+import os
+import getpass
+from pathlib import Path
+from operator import itemgetter
+from dotenv import load_dotenv
+
+# Load environment variables from .env file
+load_dotenv()
+
+import chainlit as cl
+from langchain.prompts import ChatPromptTemplate
+from langchain.schema.runnable import RunnablePassthrough
+from langchain_openai.chat_models import ChatOpenAI
+from langchain_huggingface import HuggingFaceEmbeddings
+from langchain_qdrant import QdrantVectorStore
+from qdrant_client import QdrantClient
+from qdrant_client.http.models import Distance, VectorParams
+
+# Get vector storage path from .env file with fallback
+storage_path = Path(os.environ.get("VECTOR_STORAGE_PATH", "./db/vectorstore_v3"))
+#qclient = QdrantClient(storage_path)
+
+# Load embedding model from environment variable with fallback
+embedding_model = os.environ.get("EMBEDDING_MODEL", "Snowflake/snowflake-arctic-embed-l")
+huggingface_embeddings = HuggingFaceEmbeddings(model_name=embedding_model)
+
+# Set up Qdrant vectorstore from existing collection
+collection_name = os.environ.get("QDRANT_COLLECTION", "thedataguy_documents")
+
+vector_store = QdrantVectorStore.from_existing_collection(
+    #client=qclient,
+    path=storage_path,
+    collection_name=collection_name,
+    embedding=huggingface_embeddings,
+)
+
+
+# Create a retriever
+retriever = vector_store.as_retriever()
+
+# Set up ChatOpenAI with environment variables
+llm_model = os.environ.get("LLM_MODEL", "gpt-4o-mini")
+temperature = float(os.environ.get("TEMPERATURE", "0"))
+llm = ChatOpenAI(model=llm_model, temperature=temperature)
+
+# Create RAG prompt template
+rag_prompt_template = """\
+You are a helpful assistant that answers questions based on the context provided. 
+Generate a concise answer to the question in markdown format and include a list of relevant links to the context.
+Use links from context to help user to navigate to to find more information.
+You have access to the following information:
+
+Context:
+{context}
+
+Question:
+{question}
+
+If context is unrelated to question, say "I don't know".
+"""
+
+rag_prompt = ChatPromptTemplate.from_template(rag_prompt_template)
+
+# Create chain
+retrieval_augmented_qa_chain = (
+    {"context": itemgetter("question") | retriever, "question": itemgetter("question")}
+    | RunnablePassthrough.assign(context=itemgetter("context"))
+    | {"response": rag_prompt | llm, "context": itemgetter("context")}
+)
+
+
+  
+@cl.on_chat_start
+async def setup_chain():
+    # Check if API key is already set
+    api_key = os.environ.get("OPENAI_API_KEY")
+    if not api_key:
+        # In a real app, you'd want to handle this more gracefully
+        api_key = await cl.AskUserMessage(
+            content="Please enter your OpenAI API Key:",
+            timeout=60,
+            raise_on_timeout=True
+        ).send()
+        os.environ["OPENAI_API_KEY"] = api_key.content
+
+    # Set a loading message
+    msg = cl.Message(content="Let's talk about [TheDataGuy](https://thedataguy.pro)'s blog posts, how can I help you?", author="System")
+    await msg.send()
+    
+    # Store the chain in user session
+    cl.user_session.set("chain", retrieval_augmented_qa_chain)
+
+    
+
+@cl.on_message
+async def on_message(message: cl.Message):
+    # Get chain from user session
+    chain = cl.user_session.get("chain")
+    
+    print( message.content)
+    # Call the chain with the user message
+    response =  chain.invoke({"question": message.content})
+   
+
+    # Send the response with sources
+    await cl.Message(
+        content=response["response"].content,
+
+    ).send()
+

chainlit.md

@@ -0,0 +1,6 @@
+# Let's Talk
+
+`Let's Talk` is chat app based on contents from [TheDataGuy](https://thedataguy.pro)'s blog posts.
+
+More information at [Let's Talk](https://github.com/mafzaal/lets-talk)
+

pyproject.toml

@@ -5,14 +5,17 @@ description = "Add your description here"
 readme = "README.md"
 requires-python = ">=3.13"
 dependencies = [
+    "chainlit>=2.5.5",
     "ipykernel>=6.29.5",
     "langchain>=0.3.25",
     "langchain-community>=0.3.23",
     "langchain-core>=0.3.59",
     "langchain-huggingface>=0.2.0",
     "langchain-openai>=0.3.16",
+    "langchain-qdrant>=0.2.0",
     "langchain-text-splitters>=0.3.8",
     "pandas>=2.2.3",
+    "python-dotenv>=1.1.0",
     "qdrant-client>=1.14.2",
     "unstructured[md]>=0.17.2",
 ]