init: scaffold support docs copilot backend and project structure

.gitignore

@@ -0,0 +1,21 @@
+venv/
+.venv/
+__pycache__/
+*.py[cod]
+.pytest_cache/
+.ragas_cache/
+qdrant_data/
+.env
+.env.*
+!.env.example
+.python/
+*.log
+*.err.log
+frontend.log
+backend.log
+fastembed_cache/
+.cache/
+model files
+Hugging Face cache
+Ollama models
+

README.md

@@ -0,0 +1,59 @@
+# Support Docs Copilot
+
+Python-only advanced RAG support copilot using Ollama, Qdrant hybrid retrieval, local reranking, LangGraph Self-RAG, Guardrails AI, Ragas evaluation, FastAPI, and Streamlit.
+
+## Quick Start
+
+1. Start Ollama and pull the model:
+
+```bash
+ollama run llama3
+```
+
+2. Create and activate a virtual environment:
+
+```bash
+python -m venv venv
+venv\Scripts\activate
+pip install -r requirements.txt
+```
+
+3. Ingest the sample docs:
+
+```bash
+python app/engine/ingestion.py
+```
+
+4. Start the backend:
+
+```bash
+uvicorn app.main:app --reload
+```
+
+5. Start the UI in another terminal:
+
+```bash
+streamlit run ui/app.py
+```
+
+Backend docs run at `http://127.0.0.1:8000/docs`; the Streamlit app runs at `http://localhost:8501`.
+
+## Docker
+
+```bash
+docker-compose up --build -d
+docker exec -it $(docker-compose ps -q ollama) ollama run llama3
+docker exec -it $(docker-compose ps -q backend) python app/engine/ingestion.py
+```
+
+## Suggested Commit Roadmap
+
+1. `init: setup fastapi boilerplate and environment config for ollama and qdrant`
+2. `feat: implement hybrid search ingestion pipeline with qdrant and fastembed`
+3. `feat: integrate cross-encoder reranking for context refinement`
+4. `feat: build self-rag decision graph with evaluation nodes`
+5. `feat: add input validation and output verification guardrails`
+6. `test: implement automated ragas evaluation pipeline`
+7. `feat: complete streamlit chat interface and integrate backend streaming api`
+8. `deploy: containerize complete architecture with docker compose for production`
+

data/docs/product_guide.txt

@@ -0,0 +1 @@
+An AI engineer is a specialized software professional who bridges the gap between theoretical data science and functional, real-world applications. Rather than focusing entirely on foundational model research or training algorithms from scratch, these engineers utilize pre-trained models—such as Large Language Models (LLMs) and neural networks—to build scalable, intelligent software. Their core responsibility involves creating production-ready systems, which means connecting AI capabilities to APIs, databases, user interfaces, and external tools. This highly technical role demands a strong foundation in software engineering, cloud platforms, and data management. They actively design Retrieval-Augmented Generation (RAG) pipelines, build AI agents, and implement monitoring and deployment frameworks (like Docker) to ensure models function reliably at scale. Across industries ranging from healthcare and finance to manufacturing and autonomous driving, AI engineers are essential for transforming theoretical machine learning models into robust, efficient, and user-centric products that solve complex business problems.

data/docs/sample_error.txt

@@ -0,0 +1 @@
+Error Code 404: The requested server resource was not found. Check the network router configuration in section 3.