Propagate Streamlit Space secrets to embedded uvicorn env

Default LLM to Hugging Face on Spaces; fix HF_TOKEN settings merge

Start FastAPI alongside Streamlit on Hugging Face Spaces

Fix HF Space build: bump langchain-chroma and pymupdf pins

Add Space README YAML, HF push script, and docs without binary assets

Update README and HUGGING_FACE_SPACES documentation for improved deployment guidance

Enhance Docker configuration and Hugging Face integration

Implement Docker containerization for Milestone 12

Update environment configuration and Streamlit app for enhanced timeout handling

Implement enhanced error handling for HTTP requests in Streamlit app

Enhance environment configuration and API documentation for Milestone 11

Update milestones for Milestone 10, focusing on testing and hardening the application

Enhance API and Streamlit UI with new timeout handling and telemetry adjustments

Add Milestone 9 for Streamlit UI Integration

Implement Milestone 8 by adding new endpoints for URL ingestion and collection management. Introduce `httpx` as a dependency for handling URL downloads, and enhance the API with endpoints for listing and deleting collections. Update request and response models to support new functionalities, and refactor existing routes for improved clarity and organization.

Merge branch 'main' of https://github.com/mayankchugh-learning/doc-Audi-ai

Implement Milestone 7 by introducing background ingestion jobs. Update environment configuration to include job database path, enhance job tracking with new endpoints, and refactor ingestion logic to support asynchronous processing. Modify response models to include job details and ensure proper initialization of the jobs database on startup.

Implement Milestone 6 by adding audit persistence functionality. Introduce `aiosqlite` as a dependency, update the audit routes to support event logging and retrieval, and enhance the query endpoint to persist audit records. Modify response models to include audit details and ensure proper initialization of the audit database on startup.

Update .gitignore to include audit.db and enhance milestones.md with a detailed verification flow for audit trail functionality, including backend setup and query execution steps.

Update .gitignore to exclude 'data/' directory, add platform-specific dependency pins for 'onnxruntime' and 'torch' in pyproject.toml, and modify README.md for installation instructions. Update uv.lock to reflect changes in package dependencies.

Enhance embedding and chat model integration by adding support for community Hugging Face models. Update error handling for missing dependencies and ensure API keys are securely managed using SecretStr. Refactor embedding function and chat model creation to accommodate new imports and improve robustness.

Refactor embedding and chat model integration to use langchain-huggingface. Update HuggingFace embedding function to require API key and adjust error handling. Modify retriever to import from langchain-huggingface and enhance chat model creation with temperature setting.

Update README.md to include instructions for starting the backend server with FastAPI and the frontend server with Streamlit.

Update dependencies in pyproject.toml and requirements.txt to include huggingface-hub and langchain-huggingface. Modify README.md for correct requirements installation path. Enhance uv.lock with new CUDA packages and their dependencies.

Implement Milestone 5 by adding retrieval and grounded LLM answer functionality. Update query endpoint to return grounded answers with source citations, enhance response models to include answer and sources, and refactor retriever logic to support new chunk structure and grounding mechanism. Update milestones documentation to reflect new dependencies and expected outputs.

Refactor embedding function creation and document loading. Update ingest and query routes to remove unnecessary settings parameters, streamline chunking logic, and enhance load_documents function to handle both string and list inputs. Adjust model name in embedder for consistency with OpenAI API.

ollama instalation steps

Enhance configuration and ingestion functionality for Milestone 4. Update .env.example with new environment variables, expand milestones documentation with integration details, and add necessary dependencies to pyproject.toml and requirements.txt. Refactor ingest and query routes to support file uploads and retrieval, implementing validation and error handling. Update settings to include new model configurations and parameters for chunking and embedding.

Implement Milestone 3 by adding configuration and request/response contracts. Introduce new models for requests and responses, enhance route handlers with schema validation, and update dependencies to include pydantic-settings. Update audit, ingest, jobs, and query routes to utilize new models and improve response structures.

Add Milestone 2 with route skeletons and update main.py to include new routes. Enhance health check response with app details.

Update README.md to include environment setup instructions and add requirements.txt and .env files

Initial commit