Watch the Spar Demo Video

Prerequisites: Ensure you have Docker and Docker Compose installed.

• Create a .env file in each subfolder based on the provided .env.example.
• To populate the database initially, uncomment the code in backend/app/dummy/mock_db_data.py. After the initial run, comment the code again.

Run:

docker compose -f docker-compose.dev.yml up -d --build 

to start the services.

Note: By default, this setup uses vLLM, which requires GPU support. If you're not using a GPU, disable the vllm service in your docker-compose.dev.yml.

Access applications locally at:

• Swagger API Docs: http://localhost:8000/docs
• Admin Panel: http://localhost:5174
• Sales App: http://localhost:5173

Currently using the vanilla Qwen2.5-0.5B-Instruct model.

Enabling LoRA adapters:

• Update the docker-compose.dev.yml file.
• Provide a path containing adapter_config.json and adapter_model.safetensors.

Reach out for resources and assistance:

• Anthony Malkoun: Repository and general codebase inquiries.
• Prakhar Saxena: LLM fine-tuning queries.
• Gulshan Sankhyan: UE blueprint-related questions.
• Henry Obegi: Other general inquiries.