Finish MVP and dockerize

apps/worker/worker.py

@@ -1,23 +1,68 @@
 import os
 import time
+from typing import Any
+
 from celery import Celery
 from dotenv import load_dotenv
+
+from dlib.ai import TailoringContext, generate_tailoring_suggestions
+from dlib.cv import StructuredBlock, StructuredDocument, parse_docx_bytes
+from dlib.storage import MinioStorageClient, MinioStorageConfig
+
+
 load_dotenv()
 
-# Redis connection
-redis_url = os.getenv("REDIS_URL", "redis://localhost:6379")
-app = Celery('worker', broker=redis_url, backend=redis_url)
+
+# Redis / Celery configuration
+redis_url = os.getenv("REDIS_URL", "redis://localhost:6379/0")
+app = Celery(
+    "worker", broker=redis_url, backend=os.getenv("CELERY_RESULT_BACKEND", redis_url)
+)
+
+
+def _storage_client() -> MinioStorageClient:
+    config = MinioStorageConfig(
+        bucket_name=os.getenv("MINIO_BUCKET", "resume-branches"),
+        region_name=os.getenv("MINIO_REGION", "us-east-1"),
+        endpoint_url=os.getenv("MINIO_ENDPOINT", "http://localhost:9900"),
+        access_key_id=os.getenv("MINIO_ROOT_USER"),
+        secret_access_key=os.getenv("MINIO_ROOT_PASSWORD"),
+        path_prefix=os.getenv("MINIO_PATH_PREFIX", "artifacts/cv"),
+    )
+    return MinioStorageClient(config)
+
+
+storage_client = _storage_client()
+
 
 @app.task
-def simple_task(message):
-    """A simple task that processes a message and returns a result"""
-    time.sleep(2)  # Simulate some work
+def simple_task(message: str) -> str:
+    """Basic smoke-test task."""
+    time.sleep(1)
     return f"Processed: {message}"
 
-@app.task
-def add_numbers(x, y):
-    """Simple math task"""
-    return x + y
 
-if __name__ == '__main__':
+@app.task
+def parse_document_from_storage(key: str) -> list[dict[str, Any]]:
+    data = storage_client.download_bytes(key=key)
+    structured = parse_docx_bytes(data)
+    return [block.model_dump() for block in structured.blocks]
+
+
+@app.task
+def generate_tailoring(
+    job_description: str, blocks: list[dict[str, Any]], focus_keywords: list[str]
+):
+    context = TailoringContext(
+        job_description=job_description, focus_keywords=focus_keywords
+    )
+    document = StructuredDocument(
+        version_label="worker",
+        blocks=[StructuredBlock.model_validate(block) for block in blocks],
+    )
+    suggestions = generate_tailoring_suggestions(context, document)
+    return [suggestion.model_dump() for suggestion in suggestions]
+
+
+if __name__ == "__main__":
     app.start()