AI Agent Development
Signature Course No. 202 — A Signature AI Socratic Coaching Course.
A New Kind of Learning
Socratic coaching, delivered by an agentic AI coach.
Not a course you watch. An AI coach you work with.
The Lighthouse Coach begins by discovering how you think about the subject, teaches what is missing, and gives you realistic practice before asking you to prove that you can use it. You may test out of concepts you already understand, while capstones turn the learning into a practical plan, decision, conversation, or other real-world result.
The curriculum is deliberately authored; the AI delivers it through coaching. It is not an AI randomly generating lessons.
A coach that teaches by asking — not by lecturing
The Lighthouse Coach is an AI Socratic coach built on Cognosco Learning's AI Agent Development course, in the AI at Work category. It does not play videos at you. It holds a conversation: it asks you a question, listens to your answer, names the misconception it hears, affirms what you got right, and asks the next question — the way a great coach works in the room.
An AI agent is software that does not merely answer — it acts. It reads a goal, chooses a step, calls a tool, reads the result, and chooses again, until the goal is met, the budget is spent, or a human takes the work back. Building one well is a distinct craft, and it is not the craft of writing clever prompts.
You build the system around the model. The loop the model runs inside, the contracts its tools keep, the state it carries, the limits it cannot cross, and the evidence by which you know — not hope — that it behaved: tools a model can call reliably, context and memory managed as budgeted resources, control loops that stop as reliably as they start, and a human in command at the points where command matters.
Then you take it to production. When one agent should become several, and when it should not; handoffs between agents under contract; guardrails that gate actions, not just words; evaluation that scores behaviour across a whole run; observability that lets you replay a failure instead of guessing at it; and cost, latency, security, and change budgets you set on purpose.
It works in two movements. First, Module 1 — One Agent, Built Right — work through Concepts 1–9, finishing with The Agent You Can Trust Alone; then, Module 2 — Agents in Production — work through Concepts 10–18, finishing with The System You Can Defend.
Both capstones produce real artifacts — The Agent You Can Trust Alone, a complete design and control plan for Fieldnote Systems' overnight Triage Assistant, with its pre-launch evidence list; and The System You Can Defend, a production readiness dossier for the multi-agent system — orchestration, guardrails, evaluation results, budgets, and a rollout plan, defended in writing.
The honesty discipline is part of the course. This course teaches building agents, and only building agents. Platform facts are perishable and are framed in place with a direction to check the current documentation; no statistic about agent performance in the field is asserted anywhere, and every worked figure is the running company's own, shown with its arithmetic, or a labelled supposition.
The course is self-paced and always on — roughly 30 minutes per concept, about 9 hours in full, taken in as many sittings as you like. You can type your answers, or talk out loud: the Voice Coach speaks its questions and listens to your spoken replies. Learners who already know the material can test out concept by concept.
AI Agent Development is part of The Lighthouse Signature Series — the online counterpart to the one-day, in-person learning Cognosco Learning has delivered since 2002.
Course Facts
One full concept, coached end to end — free: verify your email and we’ll send you a link to start.
Already enrolled? Enter the course →
Diagnose → Learn & Practice → Prove → Capstone
Every learner moves through the same discipline a Cognosco instructor would insist on in the room.
Diagnose
The Coach starts where the course starts: with what you actually bring. An agent is a system in which the model's output can itself cause the next event — the output selects a tool, the tool acts, and the result returns as a new observation. Every concept opens with three questions that locate where you actually stand, and your answers shape how deeply the Coach works the concept with you. Everyone begins with Module 1 — no skipping the foundations you only think you have.
Learn & Practice
Work each concept in conversation with the Coach — spoken or written — then practice it against realistic situations until it holds up.
Prove
Every concept ends in proof: answer the check cold and demonstrate the skill. Know it already? Test out and take credit immediately.
Capstone
Finish with two capstones — The Agent You Can Trust Alone for Module 1 and The System You Can Defend for Module 2 — that pull the whole course together, and earn the certificate.
What You'll Learn
A loop, not a chat. What makes a system an agent — and why agency is a property of the system, not of the model — with every fault billed to the model, the harness, or the human who owns it.
Module 1 builds one agent right. Tools as contracts the model can keep, instructions, context, and state under an explicit budget, control loops with engineered exits, memory admitted on papers, grounding by rule and by check, and a human in command.
Module 2 takes agents to production. Orchestration argued seam by seam, delegation under contract, guardrails at three gates, evaluation of whole trajectories, and observability that makes any run replayable.
And it holds the production line. Cost and latency budgets recomputed from traced actuals, least-privilege security with injection broken structurally, change that is versioned, gated, and reversible — closing with the builder's standard you sign.
What you will learn
- Say what makes a system an agent — the loop, the tools, the limits, and the exits — and separate the model's job from the harness's job and the human's
- Design tools as contracts a model can keep, and manage instructions, context, state, and memory as budgeted engineering resources
- Build planning and control loops that act, observe, adjust — and stop — with a human in command at the points where command matters
- Decide between one agent and several, and prove behaviour with guardrails, trajectory evaluation, observability, and cost, latency, security, and change budgets
Everything in the course
- All 18 concepts of AI Agent Development, in 2 modules
- A starting-point check first — the Coach finds where you actually stand before teaching
- Learn, Practice, Prove for every concept — no passive video watching
- A test-out on every concept: prove you know it and move on
- The Voice Coach: talk with the Coach out loud — it speaks, listens, and replies
- A personal Learner Record of your work across the course
- Two capstone exercises: The Agent You Can Trust Alone and The System You Can Defend
- A printable session summary whenever you pause
- The Lighthouse Coach completion certificate (Signature Series)
- Your progress is saved as you go — stop anytime, resume where you left off
Built for people who build with models
Software developers — who have called a model from code, or are about to, and who need the architecture, control, and evaluation discipline that turns a demonstration into a system.
Technical leads and architects — who must decide whether a task wants an agent at all, whether it wants one agent or several, and what proof to demand before autonomy is widened.
Platform and product engineers, and builders moving from chatbots and single calls — who ship agent features inside real products, with real users, real permissions, and real bills, and who want to learn what changes when the model starts taking actions. The course assumes you can read code and reason about a running program; it assumes no particular framework and teaches none as gospel.
Prefer to learn live and in person? The same material anchors Cognosco's one-day programs — browse the in-person catalog.
Start with The Agent You Can Trust Alone
$195 — introductory price. One payment. No subscription, no password to manage.
Here is exactly what happens when you enroll:
- Pay securely with Stripe — you are taken straight into the course
- Create your learner profile and begin the Module 1 starting-point check
- Save the access key you are given — it restores your enrollment on any device
- Learn at your pace; your progress is stored in your browser on your device
For the full voice experience, use Chrome and allow the microphone when asked. Typing your answers works everywhere, all the time. Questions before you enroll? Email hello@cognoscolearning.com.
More Signature Courses are coming
AI Agent Development is Signature Course No. 202. Leave your email and we'll send you news of future Signature Courses and Lighthouse updates as they are announced — nothing else.