I am developing a Django project that is currently hosted on Vercel as my API server. I want to implement celery workers to offload long running tasks (GenAI models prompting/outputs) and then sending emails. I have gotten familiar with running Celery locally on my machine and understand that I would have my django server running and then at the same time have my celery worker (connected to redis as a broker on Redis Cloud) running using this function in the terminal:
celery -A tasks worker --loglevel=info
My question is how do you concurrently both the Django server and the celery worker on Vercel?
Vercel is designed for serverless deployments, which means:
It doesn’t support long-running processes like Celery workers.
Each function (including your Django views) runs in isolation and terminates after execution.
There’s no persistent server environment where you can keep a Celery worker running.
Instead, you could:
Host Celery workers separately on platforms like Heroku or AWS EC2, while keeping your Django API on Vercel.
Use serverless-friendly task queues like AWS SQS with Lambda or Google Cloud Tasks instead of Celery.
Implement a webhook-based system where Vercel functions trigger tasks on external services.
Utilize Vercel’s Cron Jobs for scheduled tasks, though this has limited functionality compared to Celery.
If Celery is essential, consider migrating your entire project to a platform that supports both Django and Celery, such as Heroku or AWS Elastic Beanstalk.
