Deployment and stretch goals

The capstone moment: ship the PDF chatbot live. Stretch goals turn it from "tutorial project" into "founder-able product".

You built a working bicycle. Now you go ride it in public.

Deploy on Streamlit Cloud as in Module 6.6. Add secrets for OPENAI_API_KEY. Add a requirements.txt:

streamlit
openai
chromadb
pypdf
tiktoken
python-dotenv

Stretch goals:

1. Multi-PDF library with per-document selection
2. Hybrid search (vector + BM25)
3. Reranker (Cohere Rerank or Voyage Rerank)
4. Auth via Supabase
5. Persistent user history
6. Conversational follow-ups ("based on my last question...")
7. Image extraction from PDFs (vision LLM call)
8. Cost meter in the sidebar
9. "Suggested questions" generator from indexed content
10. Multi-language support

Hybrid search example:

from rank_bm25 import BM25Okapi
docs = [c["text"] for c in all_chunks]
bm25 = BM25Okapi([d.split() for d in docs])
def hybrid(question, k=10):
    vec_top = retrieve(question, k=k)
    kw_top = bm25.get_top_n(question.split(), docs, n=k)
    return dedupe_and_rerank(vec_top, kw_top)