Taught by a practising Principal ArchitectNEXT COHORT

Build AI that
survives production.

A four-week live cohort in agentic AI engineering. Every week ships a working system — not a prototype, not a notebook. Read the release notes below.

Apply for the next cohort → See the curriculum
6 weeks18 live sessions4 systems shipped7-day refund
The curriculum

Four weeks, shipped in order.

Each week is a release. You start with the loop and end with something that can safely touch production systems.

WEEK 01
v1.0 Foundation 3 sessions

Analysis Copilot

Build the agent loop by hand so that no framework is ever magic. By Friday you have an agent that plans its own steps and defends its answers.

  • BUILDThe think → act → observe loop, written from scratch
  • BUILDTool calling with schemas, validation, and what to do when a tool lies
  • SHIPAn agent that answers data questions with citations and a cost trail
  • DEEP DIVEWhy most agents fail on step three, and how to see it happening
agent looptool callingplanningtracing
WEEK 02
v2.0 Retrieval 3 sessions

Knowledge Navigator

Retrieval that actually retrieves. Most RAG demos work because the corpus is tiny — this week is about what happens when it isn't.

  • BUILDChunking as a design decision, not a default parameter
  • BUILDHybrid search and reranking that survives real documents
  • SHIPA navigator that knows when to say “I don't know”
  • DEEP DIVEThe retrieval failure nobody logs, and how to measure it
RAGvector searchrerankingabstention
WEEK 03
v3.0 Orchestration 3 sessions

AI Engagement Team

Several agents that divide work, disagree, and still reach a decision — without looping forever or quietly burning your budget.

  • BUILDHandoffs and arbitration between specialised agents
  • BUILDTermination conditions, retries and cost ceilings
  • SHIPA multi-agent workflow that completes reliably, not occasionally
  • DEEP DIVEWhen multi-agent is genuinely the wrong answer
multi-agentorchestrationhandoffscost control
WEEK 04
v4.0 Production 3 sessions

Enterprise Action Assistant

The week every other course skips: the one where the agent touches real systems, and someone senior asks whether it's safe.

  • BUILDPermissions, audit trails and a kill switch
  • BUILDAn evaluation harness that runs before every change
  • SHIPAn assistant that passes a production design review
  • DEEP DIVEObservability, governance, and defending the design to your CISO
guardrailsevalsobservabilitygovernanceMCP
What's included

Live, small, and reviewed by a human.

◐

Live sessions

Real time, questions answered as they come up. Recorded within 24 hours if you miss one.

◉

Personal review

Your work is read by the instructor, not scored by a bot. That review is the product.

◈

Curriculum access

Every deck unlocks as you complete its class, and stays yours afterwards.

▤

Recordings & summaries

A written recap and the full recording attached to each session.

◇

A small cohort

Capped deliberately, so the instructor knows who's stuck before you have to say so.

↺

7-day refund

Full refund within seven days of the first session. No questions.

Who teaches

A practitioner, not a professor.

Rahul Sahu

Rahul Sahu

PRINCIPAL ARCHITECT · DIRECTOR, ZS

I lead the design and delivery of enterprise AI platforms — autonomous agents, multi-agent orchestration, and intelligent workflows — with governance, observability, and resilience built in by design.

Principal Architect · ZSex-Sr. Analyst · BCGEnterprise AI & multi-agent systems
FAQ

Straight answers.

Working engineers — backend, data, platform or ML-adjacent — with roughly 3 to 12 years of experience, whose organisation now expects them to deliver something with AI. It assumes you ship code professionally. It is not an introduction to programming.

No, and we won't pretend otherwise. We don't run a hiring network. What we sell is capability: by the end you'll have built and shipped working agentic systems and be able to do the same at your own organisation.

Every session is recorded and available within 24 hours with a written summary and the slide deck. That said, the live session is the product — the cohort is small so questions get answered as they arise.

Four working systems you built yourself, the curriculum decks, the recordings, and a way of thinking about evaluation, cost and governance that most teams are missing.

Yes — we invoice against a PO. If several people from one team are interested, tell us; a private cohort is usually a better fit than individual seats.

Next cohort

Seats are limited by design.

Tell us what you're building and we'll tell you honestly whether this is the right fit — including if it isn't.