Technical Track · 3-Day Weekday Masterclass

In 3 Days, Learn to Run AI Agents That Survive Production.

A code-first masterclass on observability, evals, and incident response for agentic systems, taught by an engineer with 25 years of infrastructure behind him.

Agents are non-deterministic, chatty, and expensive. This masterclass teaches you to watch them, measure them, and keep them running, on a pipeline you build yourself.

“He holds the sole-inventor patent for LLM agent systems, granted in 2023, and has spent close to two decades delivering technical training. He has hand-built AI systems for classified environments where the public internet does not exist. In this masterclass, he shares the architecture, his working methods, and the deployment blueprints drawn from completed projects, compressed into 3 days of intensive, hands-on learning.”
Prebook AI SRE

Cohorts of 10 to 25, and seats are running out. Prebooking is free; you pay by invoice only when your cohort is confirmed.

Your Dashboards Were Built for Deterministic Systems, and Agents Are Not One.

Monitoring Goes Blind

Traditional checks pass while the agent quietly returns worse answers. Non-determinism breaks the assumptions your alerts stand on.

Evals Live in Notebooks

Teams test prompts by hand and ship on vibes. Without monitors, nobody notices quality drifting until a customer does.

Costs Run Away

Token spending scales with usage and mistakes. Without budgets and alerts, the invoice becomes the incident.

No Runbook for Agents

When an agent loop wedges at 2am, nobody knows what normal looks like or how to roll back.

Bootcamp Fatigue

Hundreds of courses, the same recycled content. The trainer sat through them too, and now teaches exactly the part they all skip.

Paper Only

Certificates confirm attendance. The concepts that matter live only in the trainer's head. Here, they live in yours.

Do these sound familiar? Operability is the fix that keeps agents employed.

Yes, Make Agents Operable

The Engineer Who Has Already Hit Every Wall.

The playbook behind every masterclass was bought with years of production attempts and hard failures. The count of the wall is the lesson, and it is the first thing he teaches.

More Than 100 Students in 2026

More than a hundred students in Singapore completed this similar AI training this year. The format is proven with a live audience.

The Patent

The sole-inventor patent for LLM agent systems, granted in 2023. The legal and logical foundation of modern agents.

GPUs Before the Buzzword

18 years of compute on GPUs, back when the field was called HPC, long before AI became a marketing buzzword.

A Trainer for Over Two Decades

Teaching engineers has been the day job for more than twenty years. Training is the craft, not a side engagement.

Air-Gap and Hybrid Infrastructure

25 years across AWS, Azure, GCP, and on-premises bare metal, including air-gapped environments few engineers ever see.

10 to 25 Seats a Cohort

Small cohorts keep the help personal. Seats are running out fast.

Past students say the most valuable part is simple. You get to ask a human who has done a few things successfully before.

Reserve My SRE Seat

The Evidence Keeps Piling Up, and It Does Not Wait.

BBC, The Straits Times · Feb 2025

The bank has said it openly

DBS is phasing out about 4,000 contract and temporary roles over three years as AI takes on work previously done by humans.

CNA · Aug 2026

Experience alone is no longer a shield

About 4 in 10 of those affected by the GovTech cuts were officers in their 40s.

Google, Fortune · Apr 2026

The curve is nearly vertical

75 per cent of all new code at Google is now AI-generated, reviewed and approved by engineers.

IBM Institute for Business Value

This hiring cycle, not 2035

40 per cent of the workforce will need reskilling within three years because of AI.

Dark Daily · Oct 2025

The scan doctors got busier, not redundant

AI now helps read CT scans, yet imaging demand grew even faster and radiologist vacancies stayed high. When AI makes something cheaper, the world orders more of it.

BBC, Business reporting · May 2025

The fintech that rehired its humans

Klarna let an AI assistant do the work of 700 service agents, then rehired people when quality slipped. The winners are the humans who can direct the AI.

None of these companies failed. They restructured, and loyalty no longer shields anyone. PwC’s 2026 jobs barometer finds roles professionalised by AI growing twice as fast as the rest. Be the one directing the machines, not the one replaced by them.

Hold My Priority Spot

Three Days. Watch It, Measure It, Keep It Alive.

Day Morning Afternoon Focus
Day 1 You study how agentic systems actually fail, and why non-determinism changes everything. The morning builds the observability mindset for AI. The afternoon has you tracing and logging real agent runs, then building the dashboard you wish you had during the last incident. A New Kind of Production System
Day 2 LLM-as-judge becomes a production monitor, and you assemble benchmark suites that run on every release. You add drift detection, set cost and latency budgets, and wire alerting that distinguishes noise from real degradation. Evals as Monitors
Day 3 You practise incident response for agents, from rollback strategies to CI gates that stop bad releases before users see them. The capstone is a monitored, gated agent pipeline you built, plus the runbook your on-call rota will actually use. Incidents and Gates

Each day comprises approximately 25% guided sessions, 60% hands-on labs on a real task, and 15% structured reflection. The table above is an example. Actual content includes more practical material than described, may differ slightly as it tracks the latest industry trends, and is refreshed right up to each cohort. The exact setup guide will be shared one week before Day 1.

Before You Arrive

What to bring for the course

  • A laptop that is ready to code, with a good CPU, sufficient memory, and free disk space for the hands-on labs
  • Administrator rights on your login account, because you will install the tools yourself
  • Everyday computer fluency; a developer background helps but is not required
  • A working knowledge of Python and JavaScript to make the labs smoother

What is provided

  • Basic AI credits for the course labs, so there is nothing to purchase beforehand
  • The professional setup guide, shared one week before Day 1

Save My Seat Now

Two Ways to Take the Masterclass.

Standard

An open cohort on the published outline

  • An open cohort of 10 to 25
  • The published 3-day outline, weekday, Mon to Fri
  • All AI credits for the labs, and the setup guide
  • The 30-day follow-up question-and-answer clinic
Prebook Standard

Custom

A private cohort tailored to your organisation

  • A private cohort for your organisation
  • A needs analysis call with the trainer before the course
  • An agenda tailored to your industry, data, and toolchain
  • Labs run on your own documents and codebase
  • A two-week post-course implementation clinic
  • Travel and venue billed separately where applicable
Enquire About Custom

Prebooking is free. You pay by invoice only when your cohort is confirmed.

What You Walk Out With

Observability for AI

Tracing and logging for agent runs, built by you on a real system.

A Live Dashboard

The dashboard that shows agent health at a glance.

Evals That Monitor

LLM-as-judge suites running in CI on every release.

Budgets and Alerts

Cost and latency budgets with alerting that catches real drift.

Incident Response

Rollback strategies and an incident process written for agents.

A Runbook That Works

A monitored, gated pipeline and the runbook that keeps it alive.

Walk out with an agent you can page someone about, confidently.

Start Operating in 2027

One Foundation. Six Masterclass Families.

The Foundation Course is the shared entry, running every weekend. Each masterclass below runs as a 3-day weekday specialist course, in a Standard open cohort or a Custom private cohort for your organisation.

Manager Track

Leader Track

You bring the domain expertise. We bring the engineering discipline. Reliability is the moat.

Prebook My Seat Now