A scalable blueprint for product customization with Python 3
summary
A concrete recipe for moving a PHP SaaS to Python 3.6 using Flask: replace per-project apps with an application dispatcher behind Nginx + Gunicorn so requests load either the core app or a lightweight custom app that copies init+config only from a template, reducing memory vs full per-instance deployments. CI/CD runs on GitLab with tests catching syntax errors. Runs ~50 events/month with 5–10 customized instances and reports the pattern works in production.
AI-written summary · about
chapters (19)
- 0:00Introduction and speakers
- 1:07About Jublia and customization
- 5:17Current PHP architecture overview
- 7:50Problems with PHP system
- 9:41Python customization challenges
- 11:30Application dispatcher solution
- 15:04Dispatcher implementation details
- 16:54Core app structure
- 18:00Customization process
- 20:29Deployment with GitLab CI
- 22:18Handling Python syntax errors
- 24:30Live coding demo setup
- 25:17Core app demonstration
- 26:41Making customization changes
- 33:26Testing and deployment
- 34:51Benefits of the design
- 35:33Migration lessons learned
- 36:41Handling product customizations
- 38:07Scalability and maintenance
related talks
36mA Python Takeover
10mThere's a Snake in the Birdhouse! Building a Python Culture
26mProtoype for Production
43mMake Your Python Web App Fly Around the World with CDNs
50mBest Practices for Python in the Cloud
36mA Python Takeover
42mBuilding a hosting platform with Python
26mDeploy Python Apps on Kubernetes and Prepare for Scale