docker-compose.yml

# =============================================================================
# docker-compose.yml — Fintel Platform
# =============================================================================
# Phase 1 services (active): postgres, pipeline, notebook
# Phase 2 services (active): chromadb, neo4j
# Phase 3+ services (commented): redis, mlflow, api, frontend

networks:
  fintel_net:
    driver: bridge

volumes:
  postgres_data:
  labels_data:
  outputs_data:
  logs_data:
  notebook_work:
  huggingface_cache:
  chroma_data:
  neo4j_data:
  neo4j_logs:

services:

  # ── PostgreSQL ──────────────────────────────────────────────────────────────
  postgres:
    image: postgres:16-alpine
    container_name: fintel_postgres
    restart: unless-stopped
    environment:
      POSTGRES_USER:     ${POSTGRES_USER}
      POSTGRES_PASSWORD: ${POSTGRES_PASSWORD}
      POSTGRES_DB:       ${POSTGRES_DB}
    volumes:
      - postgres_data:/var/lib/postgresql/data
      - ./db/init:/docker-entrypoint-initdb.d
    ports:
      - "5432:5432"
    networks:
      - fintel_net
    healthcheck:
      test: ["CMD-SHELL", "pg_isready -U ${POSTGRES_USER} -d ${POSTGRES_DB}"]
      interval: 10s
      timeout: 5s
      retries: 5

  # ── ChromaDB (vector store for RAG) ─────────────────────────────────────────
  chromadb:
    image: chromadb/chroma:latest
    container_name: fintel_chromadb
    restart: unless-stopped
    volumes:
      - chroma_data:/chroma/chroma
    ports:
      - "8000:8000"
    networks:
      - fintel_net

  # ── Neo4j (knowledge graph) ──────────────────────────────────────────────────
  neo4j:
    image: neo4j:5.18-community
    container_name: fintel_neo4j
    restart: unless-stopped
    environment:
      NEO4J_AUTH: neo4j/${NEO4J_PASSWORD}
      NEO4J_PLUGINS: '["apoc"]'
      NEO4J_dbms_security_procedures_unrestricted: apoc.*
      NEO4J_dbms_memory_heap_initial__size: 512m
      NEO4J_dbms_memory_heap_max__size: 1G
      NEO4J_dbms_memory_pagecache_size: 512m
    volumes:
      - neo4j_data:/data
      - neo4j_logs:/logs
    ports:
      - "7474:7474"   # Neo4j Browser UI
      - "7687:7687"   # Bolt protocol (driver connection)
    networks:
      - fintel_net
    healthcheck:
      test: ["CMD-SHELL", "wget --no-verbose --tries=1 --spider http://localhost:7474 || exit 1"]
      interval: 15s
      timeout: 10s
      retries: 10

  # ── Pipeline ─────────────────────────────────────────────────────────────────
  pipeline:
    build:
      context: .
      dockerfile: Dockerfile
      target: runtime
    container_name: fintel_pipeline
    restart: "no"
    env_file: .env
    environment:
      POSTGRES_HOST: postgres
      CHROMA_HOST:   chromadb
      NEO4J_HOST:    neo4j
      NEO4J_PORT:    7687
      TRANSFORMERS_CACHE: /cache/huggingface
      PYTHONUNBUFFERED: "1"
    volumes:
      - ./data/raw:/app/data/raw
      - ./data/processed:/app/data/processed
      - ./data/chunks:/app/data/chunks
      - ./data/clusters:/app/data/clusters
      - ./data/extracted:/app/data/extracted
      - labels_data:/app/data/labels
      - outputs_data:/app/outputs
      - logs_data:/app/logs
      - huggingface_cache:/cache/huggingface
      - ./src:/app/src
      - ./config.py:/app/config.py
      - ./run_phase1.py:/app/run_phase1.py
      - ./run_phase2.py:/app/run_phase2.py
    networks:
      - fintel_net
    depends_on:
      postgres:
        condition: service_healthy
      chromadb:
        condition: service_started
      neo4j:
        condition: service_healthy

  # ── Jupyter Notebook ──────────────────────────────────────────────────────
  notebook:
    build:
      context: .
      dockerfile: Dockerfile
      target: runtime
    container_name: fintel_notebook
    restart: unless-stopped
    env_file: .env
    environment:
      POSTGRES_HOST: postgres
      CHROMA_HOST:   chromadb
      NEO4J_HOST:    neo4j
      NEO4J_PORT:    7687
      TRANSFORMERS_CACHE: /cache/huggingface
      PYTHONUNBUFFERED: "1"
    volumes:
      - ./data/raw:/app/data/raw
      - ./data/processed:/app/data/processed
      - ./data/chunks:/app/data/chunks
      - ./data/clusters:/app/data/clusters
      - ./data/extracted:/app/data/extracted
      - labels_data:/app/data/labels
      - outputs_data:/app/outputs
      - logs_data:/app/logs
      - notebook_work:/app/notebooks
      - huggingface_cache:/cache/huggingface
      - ./src:/app/src
      - ./config.py:/app/config.py
    command: >
      jupyter notebook
      --ip=0.0.0.0
      --port=8888
      --no-browser
      --NotebookApp.token=${JUPYTER_TOKEN}
      --notebook-dir=/app/notebooks
    ports:
      - "8888:8888"
    networks:
      - fintel_net
    depends_on:
      postgres:
        condition: service_healthy
      neo4j:
        condition: service_healthy

  # ── Redis (Phase 3: job queue) ────────────────────────────────────────────
  # redis:
  #   image: redis:7-alpine
  #   container_name: fintel_redis
  #   restart: unless-stopped
  #   ports:
  #     - "6379:6379"
  #   networks:
  #     - fintel_net

  # ── MLflow (Phase 3: experiment tracking) ────────────────────────────────
  # mlflow:
  #   image: ghcr.io/mlflow/mlflow:v2.12.1
  #   container_name: fintel_mlflow
  #   restart: unless-stopped
  #   command: >
  #     mlflow server
  #     --backend-store-uri postgresql://${POSTGRES_USER}:${POSTGRES_PASSWORD}@postgres/${POSTGRES_DB}
  #     --default-artifact-root /mlflow/artifacts
  #     --host 0.0.0.0 --port 5000
  #   ports:
  #     - "5000:5000"
  #   networks:
  #     - fintel_net
  #   depends_on:
  #     postgres:
  #       condition: service_healthy