<- Back to blog
Platform Engineering Intensive

Insyt

A 6-month self-directed intensive to specialise as a platform engineer — with a specific focus on payment systems and real-time architecture. Every concept studied, every system built, every mistake made, published in public on Fridays.

// in progress26 weeksPublishes Fridays

Curriculum

Six months, 26 weeks. Each week produces a working system, a test suite, and a published essay. Linked weeks have a post live.

// month 01

Distributed Systems Foundations

  • W1CAP theorem, consistency models, partition toleranceRead ->
  • W2Storage engines — B-trees, LSM trees, write-ahead logsupcoming
  • W3Replication — leader/follower, multi-leader, leaderlessupcoming
  • W4Consensus — Raft, leader election, log replicationupcoming
// month 02

Payment Systems Architecture

  • W5Financial ledger design and double-entry accountingupcoming
  • W6Idempotency and exactly-once delivery semanticsupcoming
  • W7ACID transactions, isolation levels, locking strategiesupcoming
  • W8Distributed transactions — two-phase commit and sagasupcoming
// month 03

Real-time Architecture

  • W9Event-driven architecture and message queue fundamentalsupcoming
  • W10Kafka internals — partitions, offsets, consumer groupsupcoming
  • W11Stream processing — windowing and stateful operationsupcoming
  • W12Change data capture and event sourcing patternsupcoming
// month 04

Observability & Operations

  • W13Structured logging and log aggregation pipelinesupcoming
  • W14Metrics, Prometheus, and alerting designupcoming
  • W15Distributed tracing and OpenTelemetryupcoming
  • W16SLOs, error budgets, and incident responseupcoming
// month 05

Platform Engineering

  • W17CI/CD pipeline design and GitOps patternsupcoming
  • W18Container orchestration and Kubernetes fundamentalsupcoming
  • W19Service mesh, Envoy, and internal networkingupcoming
  • W20Developer experience and internal platform designupcoming
// month 06

Capstone & Integration

  • W21Security — auth patterns and secrets managementupcoming
  • W22Cost optimisation and capacity planningupcoming
  • W23Capstone: real-time payment processor, end to endupcoming
  • W24Documentation, teaching, and public retrospectiveupcoming

Reading list

Books and papers assigned alongside the build work. Core texts are read chapter-by-chapter in parallel with the relevant month.

// core texts
Designing Data-Intensive Applications
Martin Kleppmann
Primary reference throughout
The Site Reliability Engineering Book
Google SRE Team
Months 4–5 reference
Building Microservices
Sam Newman
Months 5–6 reference
// foundational papers
In Search of an Understandable Consensus Algorithm
Ongaro & Ousterhout, 2014
Raft consensus — month 1
CAP Twelve Years Later: How the Rules Have Changed
Eric Brewer, 2012
Week 1 companion
Dynamo: Amazon's Highly Available Key-Value Store
DeCandia et al., 2007
AP systems design
Kafka: a Distributed Messaging System for Log Processing
Kreps et al., 2011
Month 3
Spanner: Google's Globally Distributed Database
Corbett et al., 2012
Month 2 extension
// supplementary
The Art of PostgreSQL
Dimitri Fontaine
Kafka: The Definitive Guide
Shapira, Palino & Frick
Database Internals
Alex Petrov

Grading rubric

Each week is self-assessed against these criteria before publishing. 100 points total. The score appears at the bottom of every essay.

40pts
Correctness
The system or concept is accurately modelled and all claims are defensible with citations or working code.
20pts
Edge case handling
Failure modes, race conditions, and boundary conditions are identified and tested or documented.
20pts
Performance measured
Actual latency, throughput, or benchmark data is collected — not estimated.
10pts
Failure modes identified
At least one 'confident wrongness' scenario is documented: what breaks silently and why.
10pts
Written reflection
A published Friday essay that explains the concept clearly enough to teach it.
Total100 pts
Follow the series
New essays publish every Friday. Week 1 — CAP theorem — is live now.
Read the essays ->