# Chat Backend

Django + Channels API for AIML Operations chat (`llm_be/`).

## Local (uv)

```bash
cp .env.example .env
uv sync
cd llm_be
uv run python manage.py migrate
uv run python manage.py runserver 0.0.0.0:8003
```

## Docker (dev compose = bundled Postgres)

```bash
docker compose up --build
```

Prod compose (`docker-compose.prod.yml`) expects external `DATABASE_URL` and
`WEB_PORT` from server-infra secrets. See `.env.prod.example`.

## Ollama

Set `OLLAMA_BASE_URL` (default `http://127.0.0.1:11434` locally).
Deployed envs use the GPU host: `http://10.0.0.128:11434`.

## File storage

Prompt attachments and documents are stored in Postgres (`StoredFile` /
`DatabaseStorage`), not on the container filesystem.

## Deploy

Gitea Actions on `master` → `server-infra/scripts/deploy.sh --app chat_backend --env prod`.

Related: [chat_backend#6](https://git.aimloperations.com/ai_ml_operations/chat_backend/issues/6)