Document Q&A Assistant

from openai import OpenAI
import chromadb

client     = OpenAI()
chroma     = chromadb.PersistentClient(path="./chroma_db")
collection = chroma.get_or_create_collection("documents")

def embed_text(text: str) -> list[float]:
    res = client.embeddings.create(
        input=text, model="text-embedding-3-small"
    )
    return res.data[0].embedding

def query(question: str, doc_id: str, k: int = 3) -> dict:
    q_vec = embed_text(question)
    results = collection.query(
        query_embeddings=[q_vec],
        n_results=k,
        where={"doc_id": doc_id}
    )
    context = "\n---\n".join(results["documents"][0])
    sources = results["metadatas"][0]

    resp = client.chat.completions.create(
        model="gpt-4o-mini",
        messages=[
            {"role": "system", "content":
                "Answer ONLY using the context below. If unsure, say so."},
            {"role": "user", "content":
                f"Context:\n{context}\n\nQuestion: {question}"}
        ]
    )
    return {
        "answer": resp.choices[0].message.content,
        "sources": sources
    }

from fastapi import FastAPI, UploadFile, File
from fastapi.staticfiles import StaticFiles
from pydantic import BaseModel
from rag_engine import query, index_document

app = FastAPI(title="Document Q&A API")
app.mount("/ui", StaticFiles(directory="frontend"), name="ui")

class QueryRequest(BaseModel):
    question: str
    doc_id:   str

@app.post("/upload")
async def upload(file: UploadFile = File(...)):
    content = await file.read()
    doc_id  = await index_document(content, file.filename)
    return {"doc_id": doc_id, "filename": file.filename}

@app.post("/query")
async def ask(req: QueryRequest):
    result = query(req.question, req.doc_id)
    return result