10 parts · 17 chapters

Infrastructure and Platforms at Scale

The Cloud Engineering course covered what the cloud is made of. This course covers how to run a lot of it with many teams, without the infrastructure becoming the bottleneck. At small scale, one engineer with a console can run everything. At large scale, every resource is code, every change is reviewed and rolled out progressively, Kubernetes schedules thousands of workloads, and a platform team builds paved roads that product teams choose to walk.

Ten parts. The service layers and where responsibility shifts; infrastructure as code with Terraform and its peers, state, modules, drift and review; Kubernetes properly, from the control loop to operators; configuration and secrets across environments; CI/CD at scale with caching, artefacts and progressive delivery; platform engineering as a product; and the organisation that scales with all of it: team topologies, Conway's law, on-call and growth paths. Three further parts follow: shipping the frontend (immutable assets, atomic deploys, previews and version skew), the npm supply chain, and multi-region traffic management.

Start reading · Part 0 Present as slides
service layers · infrastructure as code · Kubernetes · config and secrets · CI/CD · platform engineering · teams · shipping the frontend · the npm supply chain · multi-regionsenior → staff · engineers who run or depend on shared infrastructure; future platform and tech leads
IaaS to SaaSThe layers, what each hands to the provider, and what that means for your team.
infrastructure as codeTerraform, CDK, CloudFormation, Pulumi; state, modules, drift, review and policy.
Kubernetes, properlyThe control loop, pods, deployments, services, ingress, scheduling, autoscaling, networking, storage, operators.
config and secretsEnvironments, promotion, drift, and rotation without a deploy.
CI/CD at scalePipelines as code, caching, artefacts, and progressive delivery.
platform engineeringThe internal platform as a product: paved roads, golden paths, developer portals.
teamsTeam topologies, Conway's law, on-call, hiring, growth paths, and the platform team's customers.
00

IaaS, PaaS, SaaS, FaaS

The layers, and where responsibility shifts
1 ch · ~10 min01

Infrastructure as Code

State, plan, review, apply, drift · Modules, environments, state splitting and the tools
2 ch · ~18 min02

Kubernetes, Properly

The control loop: objects, controllers, scheduler, kubelet · Services, ingress, networking and storage · Resources, probes, autoscaling and operators
3 ch · ~25 min03

Configuration and Secrets

Environments, promotion, GitOps and drift · Secrets at scale: distribution, rotation and response
2 ch · ~16 min04

CI/CD at Scale

Pipelines as code, caching and artefacts · Progressive delivery
2 ch · ~18 min05

Platform Engineering

Paved roads, golden paths and the portal · Running the platform like a product
2 ch · ~16 min06

Scaling and Managing Engineering Teams

Team topologies and Conway's law · On-call, hiring, growth paths and the platform's customers
2 ch · ~15 min07

Shipping the Frontend

From build output to every browser
1 ch · ~11 min08

The npm Supply Chain

Other people's code, on purpose
1 ch · ~10 min09

Multi-region and Traffic Management

Where requests go, and what each region may do
1 ch · ~10 min
Built on Cloud EngineeringThis course assumes the Cloud Engineering course: VPCs, compute, containers, IAM and secrets. It leads into SRE (keeping all of this reliable) and pairs with the Big-company FE course, whose monorepos, release trains and platform teams are the frontend half of the same story.