DevOps & Cloud
Developer on AWS
Build and ship an application on AWS properly - account setup, the compute decision, auth, databases, secrets, async patterns, idempotency, observability, testing and cost-aware design.
Developer on AWS
You know how to write the application. This course is about everything around it - the decisions you make once the code has to run on AWS and stay up.
Fourteen topics, each one a decision rather than a feature tour:
- Getting set up properly - a production-grade account from day one, how the SDK actually behaves (credentials chain, retries, pagination), and why a local development environment lies to you
- Choosing your compute - the honest decision tree across Lambda, Fargate, EC2 and EKS, then serverless and container architectures end to end
- Application building blocks - authentication and authorisation, the four-way database decision, and secrets and config the twelve-factor way
- Distributed application patterns - queue, topic, event bus and workflow, and the truth about idempotency, retries and exactly-once
- Running what you built - observability that answers questions at 3am, testing cloud applications, and designing to a cost model before the bill teaches you
The through-line is that most production incidents are not bugs in your code. They are consequences of a choice you made early and did not examine - a compute model that did not fit the workload, a retry that was not idempotent, a database picked before the access patterns were known.
Module 1 is free to preview. The full course is a one-time purchase with lifetime access.
What you will learn
- Set up an AWS account for production, not for a tutorial
- Choose between Lambda, Fargate, EC2 and EKS from the workload rather than the hype
- Build serverless and container applications end to end, with the failure modes named
- Get authentication, authorisation and secrets right before they become an incident
- Pick the right database from your access patterns, across the four realistic options
- Use queues, topics, event buses and workflows for what each is actually good at
- Write handlers that survive retries, and understand why exactly-once is an illusion
- Instrument, test and cost-model a cloud application so production holds no surprises
Prerequisites
- You can build a backend application in some language - this course is about running it on AWS, not about programming
- Basic familiarity with HTTP, JSON and the command line
- An AWS account is useful to follow along but not required
Course outline
Module 1: Getting Set Up Properly
Getting Set Up Properly - 3 topics.
- Your first production-grade AWS account setup 11 min read
- The AWS SDK model: clients, credentials chain, retries, pagination 10 min read
- Local development against AWS - and what lies to you 9 min read
Module 2: Choosing Your Compute
Choosing Your Compute - 3 topics.
- Lambda vs Fargate vs EC2 vs EKS - the honest decision tree 9 min read Pro lesson
- Serverless application architecture end to end 10 min read Pro lesson
- Container application architecture end to end 11 min read Pro lesson
Module 3: Application Building Blocks
Application Building Blocks - 3 topics.
- Authentication and authorisation for your application 9 min read Pro lesson
- Storing application state - the four-way database decision 9 min read Pro lesson
- Secrets, config, and the twelve-factor app on AWS 9 min read Pro lesson
Module 4: Distributed Application Patterns
Distributed Application Patterns - 2 topics.
- Async patterns: queue, topic, event bus, workflow 10 min read Pro lesson
- Idempotency, retries, and exactly-once illusions 10 min read Pro lesson
Module 5: Running What You Built
Running What You Built - 3 topics.
- Observability for application engineers 10 min read Pro lesson
- Testing cloud applications 9 min read Pro lesson
- Cost-aware application design 9 min read Pro lesson