The full AI agent course curriculum

Eight live sessions across four weeks.
Stop dabbling with AI. Start delegating to effective agents.
Leave with proactive agents that do real work for you.

Cohort 4 · kicks off Tuesday 29 September 2026Suitable for non-technical folksAdvanced lane for Claude Code users

400+ people trained so far

Cohort seat £1,000 +VAT with a 100% money-back guarantee.
Self-paced = every recording, guide, and template, in your own time - £375 +VAT.

Works with your existing AI setup

  • Claude Chat
  • Claude Cowork
  • Claude Code
  • ChatGPT For Work
  • Codex
Week 129 Sep – 2 OctYour second brain & safety
1
Tue · 29 Sep · 2:30–5:00pm UK

Agent Foundations: Your Personal Second Brain

Build in Claude or ChatGPT: as a Project, in Cowork, or a local folder for Claude Code

Everything in this programme is built on one asset: your personal second brain - a portable folder of context files about you, your work, and your goals that any agent can read. In this first session you build it, with your first purpose-built agent running on top. Most people shape it as a personal strategy coach that knows their background, their goals, and the patterns that hold them back. Every session after this one builds on it.

What you'll build

  • Your first agent, in the container that fits you: a Project in Claude or ChatGPT, a folder-backed Cowork agent, or a local folder for a coding agent
  • The seed of your personal second brain: context files that capture your work, your team, and how you think, so the agent stops giving generic advice
  • Custom system instructions generated through a guided, coaching-style interview
  • A background knowledge base, built with Deep Research, that your agent draws on every day

What you'll learn

  • How agents actually work: the brain (the model), the context (memory and files), and the tools it can reach for
  • What a second brain actually is - a portable folder of context files any agent can read - and why it beats relying on chat memory or clever prompting
  • When to use a Project, when to use Cowork, and when to reach for a coding agent
  • The art of context management: freshness, file size, relevance, and the file structures that make agents remember what matters

Going further (advanced)

  • Already using Claude Code daily with one endless chat and no context files? This is where you split that into a CLAUDE.md and a context-file structure your agent reads every session - the upgrade past technical participants said paid off most
  • Structuring the folder so the same brain works from Claude, ChatGPT, a coding agent, or whatever comes next

What you won't get anywhere else

I'll share two prompts I've refined over months of daily use: an Interview Prompt that interrogates your communication style, decision-making, and blind spots to generate deeply personalised instructions, and a Personal OKR Prompt that gets you crystal clear on what you're actually optimising for. One Cohort 3 graduate called building his second brain 'a game changing moment'; another called this session 'the biggest single breakthrough' of the programme.

2
Fri · 2 Oct · 2:30–4:30pm UK

Security

Working with agents safely

Before you wire AI into your CRM, calendar, and files, you need to know the risks. You leave with a clear mental model for what's safe to connect, what isn't, and the vetting checklist you'll use before installing any integration in Week 2.

What you'll learn

  • Prompt injection in plain English: how hidden instructions get into the content your agent reads, why email is the highest-risk channel, and how to shrink the attack surface
  • Why permissions beat instructions: instructions are guidance, not a guarantee, so scoping what an agent can see, do, and change is the single most reliable control
  • The lethal trifecta: why private data, untrusted content, and the ability to send things out should never meet in one agent, and how to keep the three apart
  • How to vet a new integration (an 'MCP server') before you trust it: official versus DIY, read-only first, and the safe-first-test pattern
  • The higher-risk territory (browser control and computer use) and the compliance questions to ask before adopting any AI tool

Going further (advanced)

  • Run a prompt injection on yourself: plant a hidden instruction in a test email and see whether your agent obeys it or catches it
  • Fish your own history: ask your agent to sweep past conversations for pasted API keys, passwords, and sensitive data, then clean up what it finds
  • On a Team or Enterprise plan: set (or request) organisation-level security instructions that apply to every seat

What you won't get anywhere else

We use real moments of confusion from past cohorts (file-sharing permissions, a live prompt-injection demo) as the teaching examples, so the risks land concretely instead of as abstract warnings.

Week 26 Oct – 9 OctYour tools & reusable skills
3
Tue · 6 Oct · 2:30–5:00pm UK

Connect Your Tools (Integrations & MCP)

Native connectors and MCP servers for your CRM, task manager, and calendar

Your agent goes from smart-but-blind to wired into your actual tools. We use a go-to-market pipeline as the worked example, but the techniques apply to any workflow: connect your CRM, task manager, calendar, or any platform so your agent can read, act, and automate across them.

What you'll build

  • Live connections between your two or three most-used tools, so your agent works from your real data
  • A multi-tool workflow, for example: find deals untouched for a week, then create follow-up tasks with full context
  • A diagram of your main platforms and where data flows between them, ready to share with your team
  • A ranked list of automation ideas for your business, sorted by impact and effort

What you'll learn

  • What an API is and what MCP (the Model Context Protocol) is, explained in plain English
  • Connectors, MCP servers, integrations, plugins: the jargon untangled, so you can pick the right way to connect any tool
  • Chaining tools together so platforms that don't talk to each other directly can work through your agent
  • The safe-first-test pattern: read before you write, scope permissions, and keep your agent fast by not drowning it in tools

Going further (advanced)

  • Build your own MCP server for a tool that has no connector, and find out why tool descriptions are really prompt engineering
  • The pass bar: a fresh session, given only your server, completes a real task without you naming the tool it should use

What you won't get anywhere else

A couple of years ago this kind of setup would have taken weeks of custom development. Now you build it in a single session. I've run this exact stack daily since late 2025 across consulting, coaching, and sales, and you'll see the real workflows live: a pipeline review that surfaces stale deals, call prep with research pulled in automatically, and post-meeting follow-ups that update the CRM and create tasks in one go.

4
Fri · 9 Oct · 2:30–4:30pm UK

Reusable Agent Skills

Named workflows you write once and run forever

Skills are the unlock that makes everything afterwards easier. You leave having shipped at least one skill, with a clear sense of what separates a genuinely good skill from an OK one, and how to share it with your team.

What you'll learn

  • The taxonomy that ends the confusion: context file versus system instruction versus skill versus plugin versus integration, and when to reach for each
  • What is inside a great skill: the description that makes it trigger at the right moment, the steps, and the examples
  • The quality moves that separate a reliable skill from a flaky one: source-of-truth rules, context discipline, and a troubleshooting block
  • A skill you install is code you're about to run: how to audit a skill from a public directory before trusting it
  • Three patterns for sharing skills across a team, from sending a single file to a maintained repository with named owners

Going further (advanced)

  • A Claude Code skill that pins its own model and reasoning effort in its frontmatter, so the job always runs on the right brain regardless of your session default
  • Scoping a skill's tool access to only what it needs, so a shared skill is safe for a colleague to run

What you won't get anywhere else

We dissect real skills I use every day - including the multi-agent refine loop that pressure-tests my own work - then build a brand-new skill live with a build-test-improve loop that compresses days of trial-and-error into minutes. Newcomers leave with a working skill they can copy; advanced builders leave with the workflow for hardening the skills they already have half-written.

Halfway through the curriculum. ~4.5 hrs of live sessions a week, all recorded. Cohort 4 kicks off Tuesday 29 September 2026.

Book your seat(s)
Week 313 Oct – 16 OctAlways-on agents & your company second brain
5
Tue · 13 Oct · 2:30–5:00pm UK

Proactive Agents

Scheduled, always-on agents

This is where your agents stop waiting for you. You leave with at least one proactive agent running on a schedule that genuinely earns its place (not a daily info dump you learn to ignore), and a clear ladder for making it more useful every week.

What you'll build

  • Your first scheduled agent that runs on its own, reads yesterday's output, and reports back without being asked
  • An upgrade path that adds your business context so it suggests actions, not just summaries
  • A stretch build where the agent drafts the work for you, not just flags it
  • A plan for where it should live, from running on your laptop to cloud-hosted and always on

What you'll learn

  • The four levels of proactive agents: information, then suggestions, then drafted work, then closing the loop
  • What makes a proactive task worth keeping: it acts on your context, respects your attention, and gets better over time
  • Real, working examples: a morning briefing, automatic capture of commitments from your meetings, and a daily revenue check
  • Where a routine runs and what fires it, and when to graduate from your laptop to an always-on, cloud-hosted agent

Going further (advanced)

  • Move your agent to the cloud so it runs with your laptop shut - and the four setup traps you'll hit on the way
  • Close the loop: give an agent its own run log so it works out what it keeps getting wrong and proposes edits to its own brief
  • The weekly workspace review: an agent that audits your skills, context files, and connectors, and tidies the folders they live in

What you won't get anywhere else

I'll walk through Jamie, my own sales agent, built over weeks of iteration, showing the real output at each of the four levels - from noisy and ignored to trusted with real work - so you can see exactly where yours can go. The skills you built in Week 2 are what make the higher levels possible.

6
Fri · 16 Oct · 2:30–4:30pm UK

Company Second Brain

Growing your personal second brain into shared context layers your whole team draws on

Weeks 1 to 3 built your agents. This session gives your whole team a brain to share. The personal second brain from Session 1 becomes the seed of a company one - layered context (personal, team, company) that every agent and every colleague can draw on. One Cohort 3 participant built a customer-support brain from 11 internal documents that helps her team resolve tickets when no experienced colleague is online; this session is how you build yours.

What you'll learn

  • The context-layers model: what stays personal, what your team shares, and what belongs to the whole company
  • The build recipe: pick one blessed source per topic, convert it to a simple text format agents read best, index it, and name a human owner who keeps it fresh
  • How to keep it alive: owner contracts with a sync promise, links back to where each fact came from, and scheduled sweeps
  • The start-solo path: build the company brain yourself first and use the working demo to win the team over - it pays off for a business of one too
  • The realistic rollout picture: you don't need everyone building, you need a focused group building context and skills everyone else benefits from

Going further (advanced)

  • The permissions rule for shared agents: minimum permissions anyone in the organisation could have, or the agent inherits the identity - and the permissions - of the human running it
  • Usage rules that grow from corrections: teach the brain when it gets something wrong, and it stops repeating the mistake - no team sign-off required
  • When knowledge lives in a database, not documents: files for knowledge, a database for numbers, and a map that covers both

What you won't get anywhere else

This is built from watching real companies do it: guest leaders who have opened up their actual company brains inside the programme, from a 200-person scale-up to companies past 1,000 people, plus around twenty company builds from the last cohort. You see what worked and, just as honestly, what failed - including why exporting files to your team goes stale, and what the companies that cracked it do instead.

Week 420 Oct – 23 OctBuild software & showcase
7
Tue · 20 Oct · 2:30–5:00pm UK

Coding Agents

Claude Code, Cowork, Codex, or Cursor (your choice)

Write a clear, one-shot brief, hand it to a coding agent, and watch it build. No coding required. You experience directing an agent that already knows how to code, and the power of running several in parallel. Software teams ship a working prototype; everyone else ships a validation page to test demand.

What you'll build

  • A one-shot PRD (product requirements document): a brief clear enough to hand to a developer, an agency, or an AI agent and let them build
  • A live, deployed prototype or validation page, built from your brief
  • Something you can share with users, your team, or investors the same day

What you'll learn

  • The mindset shift: you're not learning to code, you're learning to direct an agent that already can
  • The seven-step build process: from idea, through deep research, to a one-shot PRD with milestones - and why the plan lives in a file the agent ticks through, not in the chat
  • Which tool fits which job: Cowork versus Claude Code versus Codex for custom builds, and where quick-prototype tools like Lovable and Replit fit
  • Directing an autonomous agent: when to let it run, when to step in, mocking up the visual first, and handling API keys and deployment safely

Going further (advanced)

  • Run more than one agent in parallel, each ticking through its own milestone of the PRD
  • Keep a long build on track when it outgrows the agent's memory: milestones in a file, so a fresh session picks up exactly where the last one stopped

What you won't get anywhere else

I build with these agents daily - prototypes, internal tools, landing pages, and full applications, including the platform this course runs on. You get the distilled version: which tool to use when, the prompting patterns that actually work, the mistakes that waste an afternoon, and the shortcuts that save one. Guided building with someone who has done enough of it to know where things break.

8
Fri · 23 Oct · 2:30–4:30pm UK

Showcase

Demo day

The finale. Members of the cohort demo what they built (demoing is optional, and you choose what to share), we celebrate the standout agents from across the four weeks, and we map out what to build next. In Cohort 2, 24 people demoed working agents four weeks after starting - many of them non-technical when they began.

Book your seat for Cohort 4 - kicks off Tuesday 29 September 2026 →

What's included

4 weeks • 8 live sessions (~4.5 hrs a week, UK time) • weekly guest speakers & office hours • yours to keep

4× Live Build Workshops

Tuesdays · 2.5 hrs · UK time

Each week you build a new working agent on top of your second brain, live with the cohort. An hour of teaching, then a guided co-build.

3× Tactics Sessions + Demo Day

Fridays · 2 hrs · UK time

Sharper craft on top of the week's build - security, skills, the company second brain - then the final Friday is demo day.

Checkpoints & Certificate

Weekly build checkpoints keep you shipping (with an advanced lane on most), and completing the programme earns a certificate.

Guest Speaker Lunch & Learns

Thursdays

A weekly session with a specialist guest, going deeper on a topic that complements the week. Previous speakers have come from AI-native companies including Fin (Intercom), Incident.io, Jack & Jill and Rig.

Weekly Office Hours

Grouped by department

Drop-in sessions by function to get feedback on your agents and see how others are solving the same problems.

Private Coaching Calls (Add-on)

Not included - optional extra

Private consulting and coaching calls with James and his coaches are available throughout the programme, on your specific agents and AI rollout.

Private Community

Four weeks inside a group with James and the cohort. Share wins, swap notes, and get unstuck between sessions.

Full Recordings & Guides

Every session is recorded, with step-by-step build guides, so you can catch up or go deeper any time.

Prompts, Templates & Frameworks

The mega-prompts, system instructions, skills, and agent frameworks you build are yours to keep and adapt.

Partner Perks

Discounts on Granola, Attio, and more.

Works with Claude, ChatGPT, and whatever AI tools you already use.

Frequently asked questions

Do I need to be technical to take the Agent Accelerator?

No. The course is designed to work for non-technical operators: if you've used ChatGPT, you're ready. Every session offers a build container to match your comfort level, with plain-English guides throughout - many graduates began with nothing more than ChatGPT experience.

I already use Claude Code every day - will I learn anything?

Yes - daily use is not the same as leverage. Most daily users run one endless chat: no CLAUDE.md, no context files, no skills, nothing scheduled. The course is where that changes. Every session has an advanced lane with its own optional stretch checkpoints on the platform: building your own MCP servers, skills that pin their own model and effort, cloud routines that run with your laptop shut, and agents that read their own run logs and improve their own briefs.

How much time does it take each week?

Around 4.5 hours of live sessions a week - Tuesdays 2:30–5:00pm and Fridays 2:30–4:30pm, UK time - plus optional guest talks, drop-in Q&A sessions and office hours, and build time between sessions. Every session is recorded, so a missed week does not derail you.

What do I need before starting?

A paid Claude or ChatGPT plan (£20 a month minimum; the free tiers won't get you through the course) and a laptop you can install integrations on. If your company restricts AI tools, there are read-only and no-connector paths through the course, and the Week 1 security session covers working within your organisation's AI policy.

What is a second brain?

A portable folder of context files - who you are, how you work, what you are aiming at - that any AI agent can read. You build a personal one in Session 1, every later session builds on it, and in Week 3 you grow it into a company second brain your whole team can draw on.

How much does the AI agent course cost?

A seat on the live four-week AI agent training (Cohort 4) is £1,000 +VAT, with a 100% money-back guarantee: complete the programme, attend the sessions (recordings count) and complete the checkpoints, and if you are not satisfied with the value, we refund in full. The self-paced course - every recording, guide, and template - is £375 +VAT. Teams of 10 or more get volume pricing (£900 +VAT per seat).

Is there a certificate?

Yes. Weekly build checkpoints track your progress through the programme, and completing it earns a certificate.

How current is the curriculum?

It is rebuilt for every cohort, and it has to be. The tools change too fast for a recorded course to stay useful for long: MCP, agent skills, scheduled agents and coding agents like Claude Code and Codex all either did not exist or worked very differently a year ago. So before each cohort James goes back through every session and rebuilds it around what people are actually using now, drops anything that has been superseded, and adds the new workflows he is running in his own business that week.

Sessions get replaced between cohorts as the leverage moves, the security session is updated as new attack patterns appear, and because James builds and runs agents daily (including the platform this course runs on), the examples you see are live workflows, not slides from six months ago.

The self-paced course gets the same treatment. When a cohort finishes, its new recordings and guides replace the previous ones on the platform, and everyone with access gets the update for free. Whichever option you choose, you are learning the current version of the craft.

Can I buy the course for my team?

Yes. Teams of 10 or more get volume pricing (£900 +VAT per cohort seat), and we can invoice your company. Email hello@jamesmcaulay.co.uk to set it up.

Ready to build?

4 weeks. 8 live sessions. Agents that take real work off your plate - the follow-ups, the briefings, the busywork - and give you hours back every week.

Cohort 4 kicks off Tuesday 29 September 2026 · seat £1,000 +VAT · 100% money-back guarantee

Book your seat(s)Buy self-paced course →

Self-paced is every recording, build guide, and template from the programme - £375 +VAT.
Buying for a team of 10+? Book a discovery call.

Questions? Email hello@jamesmcaulay.co.uk