Building Serverless applications with Python
summary
Practical walkthrough of deploying Python on AWS Lambda using Chalice, Zappa and raw AWS CLI with concrete limits and gotchas: OpenCV-based face detection is built, deployed and returns rectangles but the dependency zip is ~34 MB (zipped limit and 500 MB unzipped constraint force S3 uploads), debugging relies on local event/context mocks, Chalice creates a single Lambda and leaves CloudFormation resources behind, API Gateway lacks WebSocket support so real-time requires IoT or custom workarounds.
AI-written summary · about
chapters (18)
- 0:00Introduction to serverless
- 2:40History of serverless computing
- 4:07How AWS Lambda works
- 5:35Creating a Lambda function
- 7:05Triggering Lambda with API Gateway
- 8:13Introducing Chalice framework
- 10:01Deploying with Chalice
- 11:12Infrastructure management with SAM
- 12:16AWS Lambda limits and free tier
- 13:22Zappa for Django applications
- 14:30Exploring Lambda runtime environment
- 16:57Python 2.7 vs Python 3 in Lambda
- 17:41Adding dependencies to Lambda
- 18:49Future of serverless frameworks
- 20:14Handling large dependencies
- 22:03Debugging Lambda locally
- 23:30WebSockets and Lambda limitations
- 24:37Q&A and closing remarks
related talks
25mA Crash Course on Serverless Applications in Python
34mPython in Serverless Architectures
30mLessons from building serverless services in bleeding edge python
19mBuilding Serverless API Backends
54mServerless Python on AWS Lambda using the Serverless Framework
50m Serverless Python in the real world: common patterns & challenges 
29mGoing serverless with Django on AWS Lambda
40mBest Practices for Deploying Well-Architected, Serverless Python Applications to AWS with CDK