Proper incident drills, actual war-room exercises, real config puzzles — not pre-recorded videos you'll half-watch at 2× speed. We built these courses because we kept seeing the same mistakes on the same outages, and slide decks weren't fixing anything.
2,180+
Trained
96%
Trial-to-Paid
152
Lab Scenarios
8 min
Avg Response
We don't organise by vendor or product name. Each course is built around a specific failure pattern we've seen hit production — because that's the thing that actually matters at 3am on a Saturday.
Your pager goes off. Twelve people join the call. Nobody knows who's in charge and everyone's running different queries against different dashboards. Sound about right? This course drills the actual workflow — role assignment, communication cadence, timeline tracking — using fault-injected labs that break in ways you'll recognise from your own post-mortems. By week three, most teams have cut their mean-time-to-acknowledge in half, though honestly some of that is just people finally agreeing on which channel to use.
Start TrialThe longest course we run, because migrations are where careers go to die — or at least where weekends go to die. You'll practise blue-green cutover, canary routing, data-sync verification, and rollback under pressure. Every lab has a "surprise" constraint: one gives you a 2GB memory ceiling, another drops network between zones mid-migration. We added those because, well, that's what actually happens. Ten weeks sounds like a lot, but the teams that finish this one stop scheduling Saturday maintenance windows.
Start TrialNot a security course for security people — this is for the ops engineer who just got told "you also own hardening now" and isn't sure where to start. Covers baseline configs, secret rotation, network segmentation, and audit logging. The labs let you run an actual scanner against your own provisioned environment and then fix what it finds, which turns out to be a much better teacher than reading a 400-page compliance PDF. Fair warning: the vulnerability scanners in lab 6 are, ah, thorough. Budget extra time for that week.
Start TrialDashboards everywhere and still no one saw the outage coming? Yeah. This course is about building monitoring that actually tells you something useful — alert thresholds that aren't just "CPU > 80%", capacity models you can show to finance, and runbook automation that works at 3am when the on-call engineer is running on two hours of sleep. The final lab is a simulated traffic surge where you have to predict the failure point before it happens. Most people get it wrong the first time. That's rather the point.
Start TrialWe're not saying every vendor course is bad — some are fine. But the structure of most training programmes hasn't changed since about 2009, and the problems definitely have.
| Feature | Lockhart Ops | Typical Vendor Course |
|---|---|---|
| Lab environment | Live multi-node clusters, fault injection, your own SSH access | Shared sandbox, pre-configured, resets every session |
| Scenario design | Based on real outage patterns from post-mortems | Based on product documentation chapters |
| Instructor access | Direct message any time, 8-minute average reply | Office hours once per week, email queue |
| Team exercises | War-room simulations with role rotation | Individual quizzes and multiple-choice exams |
| Accreditation | Not government-accredited — we're honest about that | Often vendor-certified (recognised by that vendor only) |
| Pricing | £1,800–£2,650 per course, no hidden fees | £2,000–£4,500+ with exam fees, retake fees, annual renewals |
62% faster
"Mean-time-to-resolve dropped from around 47 minutes to 18, which — honestly I didn't believe the numbers at first, I thought someone had misconfigured the tracker. But no, it held. Three months later and we're still sitting at about 20 minutes on average. The war-room drills changed how the team communicates during incidents, full stop."
— R. Anwar, Platform Lead, 90-person SaaS company
0 weekend deploys
"We used to do all major releases on Saturdays because we were terrified of breaking things during business hours. After the zero-downtime course, we moved everything to Tuesday afternoons. It's been — what, five months? — and we haven't had a single rollback. My team actually has weekends now, which sounds dramatic but it genuinely changed morale."
— S. Kapoor, Ops Manager, financial services
14 to 3 alerts/night
"Before the observability course we had something like 14 alerts hitting the on-call phone per night, most of them noise. Afterwards we rebuilt our alert rules from scratch — which the instructors basically forced us to do, and I'm glad they did — and got it down to 3 actionable ones. The on-call rotation went from something people dreaded to something that's, well, fine."
— D. Okonkwo, Senior SRE, e-commerce platform
Fourteen days, full access, no credit card. If the labs don't convince you, the trial just ends — no follow-up calls, no guilt trips, no "special offer if you stay."
You get full access to every lab scenario we have — all 152 of them. No restrictions, no "lite" version. You can spin up environments, break them, fix them, repeat. We don't ask for a credit card and there's no automatic billing. On day 12, we send a reminder that the trial's ending. On day 14, access stops. That's it. If you want to enrol in a full course after that, great. If not, you don't owe us anything and we won't pester you about it.
No. We're not government-accredited and we don't pretend to be. You won't get a certificate you can wave at a regulator. What you will get is hands-on experience with realistic failure scenarios, which — in our admittedly biased opinion — matters more when you're staring at a broken cluster at 2am. We issue a completion certificate with your name and course details on it, but it's from us, not from a government body. Some employers count it towards CPD hours; check with yours.
Roughly 6–8 hours per week for most courses. Some of that's structured lab time, some is working through scenarios at your own pace. The migrations course (10 weeks) runs a bit lighter at around 5–6 hours per week because it's spread over a longer period. We've had people cram it into less, but they tend to get less out of the labs. The fault-injection exercises need time to sit with — you don't want to rush a scenario where the whole point is learning to stay calm under pressure.
Yes, and frankly that's how the courses work best. The war-room exercises in particular require multiple people playing different roles — you can do them solo, but it's a bit like practising a fire drill by yourself. We offer group pricing for teams of 4 or more: ring us on +44 121 647 3582 or email enrol@lockhartops.co.uk and we'll sort out a quote. Each person gets their own lab environment either way.
If you cancel 14 or more days before your course starts, you get a full refund. Between 7 and 13 days out, it's a 50% refund. Once the course has started, we can't offer a refund — but you can defer your place to a later cohort once, at no extra charge. If you need to re-enrol after that, there's a £225 re-enrolment fee. Full details in our Terms of Service.
Message an instructor directly — our average reply time is about 8 minutes during UK business hours (09:00–18:00 Mon–Fri). Outside those hours it's more like 30 minutes, depending on who's on call. We don't use chatbots or auto-responders. You'll get a real person who's done the lab themselves. If you're properly stuck, they can join your lab session and walk through it with you. We'd rather do that than have you waste two hours going in circles.
14 days. 152 labs. No credit card. No automatic billing. We'll send a reminder on day 12 so the expiry doesn't catch you off guard.