Designed around real postmortems, because that is where the lessons are.
Watch a free previewOffer ends Sunday at midnight
30-day refund, no questions asked
Training a team?
Volume pricing from five seats, with assignment, reminders and completion reporting included.
See team plans →Each maps to the same taxonomy your organization’s roles are defined in, so progress here moves your skill profile.
Infrastructure taught incident-first. Each module opens with a failure — lightly anonymised from real production systems — and works backwards to the architectural decision that allowed it. Covers distributed systems fundamentals, Kubernetes, observability and the operational practices that make on-call survivable.
4 modules · 16 lessons · 10h 2m
Starting from what goes wrong.
Arjun has been on call for production systems for eleven years and has the scar tissue to prove it. He teaches infrastructure the way he wishes it had been taught to him: incident first, architecture second. His courses are built around real postmortems, lightly anonymised, and he will spend an entire lesson on a single failure if the failure is interesting enough.
3,894 reviews
Rolled this out to fourteen people. Twelve finished, which for us is unheard of. The pacing is right and nobody complained about the length, which is the real test.
Solid content, but I came in from a non-technical background and the assumed knowledge in module four caught me out. The prerequisites could be clearer about that. Everything before it was excellent.
Genuinely good, with one caveat: the first module moves slowly if you already know the basics. Skip to module two if you do — the rest is excellent and the case studies are specific rather than generic.
The engineering around the model, which is where projects actually fail.
Anders Lindqvist · Arjun Mehta
18.9K
Types that catch real bugs without making the codebase unreadable.
Daniel Roth
62.8K
Measure, then fix — in that order, every time.
Daniel Roth · Ben Kowalski
47.1K