11Aug
Tue
Session 01
All cohorts
AI Agents
Open for registration
AI Agents
Foundations.
Turn AI into your ultimate competitive advantage for your career and business.
Stop drowning in repetitive work. In two weeks, direct AI to automate your most exhausting tasks and surface high-value opportunities. No coding needed.
Build loop
Define
01 / 06
6 practical stages.
The outcome
What you'll
walk away with.
Three working AI agent workflows you have built, tested and can run yourself:
1. A Reliable Document Skill that drafts to your standard.
2. A Document Desk Controller that processes multiple files safely and records what happened.
3. A Sourced Opportunity Scout that finds and ranks results with evidence attached.
You also leave with a repeatable method for designing, evaluating and improving future workflows.
What you'll learn
Syllabus.
01
AI Agent Foundations and Build-or-Buy Strategy
Understand how AI agents work and decide when to use, buy or build one.
You will understand what separates an AI agent from a chatbot, a prompt and a fixed automation. You will assess real workflows, identify where an agent can create value, and decide whether to use a ready-made tool, configure an existing platform or build a custom workflow. You will also define the outcome, boundaries, permitted data and human approval points for your first agent.
02
Prompt Engineering and Instruction Design
Write precise instructions that make AI agents more reliable and useful.
You will learn how to define the goal, source materials, operating rules, output structure and failure behaviour of an agent. You will design instructions that help the agent recognise missing information, ask the right questions and produce consistent results instead of confident guesses. You will apply these principles while creating your Reliable Document Skill.
03
Context Engineering
Control what the agent knows, uses and pays attention to during its work.
You willl learn how to select and structure the documents, examples, rules and reference materials an agent needs to perform well. You will distinguish permanent instructions from task-specific context, manage conflicting information and prevent irrelevant material from weakening the result. Your Document Skill will use this context to produce work that follows your standards.
04
Agent Architecture and Workflow Orchestration
Design multi-step agents that use tools, make decisions and complete work safely.
You will learn how inputs, tools, decision rules, outputs and approval points connect inside an agent workflow. You will explore controllers, batch processing, exception handling, run records and human-in-the-loop design. You will apply these concepts by building a Document Desk Controller that processes multiple files without overwriting originals or hiding failures.
05
Agent Evaluation, Reliability and Safety
Test an agent systematically and prove that an improvement actually works.
You will learn how to create normal, incomplete and difficult test cases, define observable quality criteria and establish a performance baseline. You will make focused improvements and rerun the same tests to confirm that quality increased without breaking something else. You will also cover data boundaries, tool permissions, unsafe instructions and the points where human judgement must remain in control.
06
Research Agents and Evidence Engineering
Build research agents that find, verify and rank information with evidence attached.
Learn how to design bounded research workflows that preserve source links and dates, separate eligibility from ranking, remove duplicates and mark missing evidence as unknown instead of guessing. You will apply these techniques by building a Sourced Opportunity Scout for grants, tenders, leads, partners, roles or other opportunities. You will finish by demonstrating one build and scoping a practical pilot for your work, organisation or first client.
When the cohort runs
Schedule.
Six live sessions of ninety minutes, three evenings a week across two weeks. Tuesdays, Thursdays and Saturdays, 6:00 to 7:30 PM WAT. Five sessions run 90 minutes and the Opportunity Scout lab runs 120 minutes. Every session is recorded and shared afterwards, so a missed session is easy to catch up on. Because the course is cumulative and assessed, certificate candidates must still attend at least five live sessions, including the Scout lab or its guided clinic.
13Aug
Thu
Session 0215Aug
Sat
Session 0318Aug
Tue
Session 0420Aug
Thu
Session 0522Aug
Sat
Session 06Completion
What completion
looks like.
To complete the cohort and earn your certificate, you take part in at least five of the six live sessions, build and run your own agent through the weekly assignments, and present it on demo day.
Cannot make demo day live? A short recorded demo counts instead. Completion here means something concrete: you walk away with a working agent you built, not just notes.
5/6sessions
01Attend
02Build
03Test
04Demonstrate
Certificate earned
Before you start
Before you join.
No coding background needed. You need a laptop on a current macOS or Windows, a stable internet connection, the latest Claude Desktop with a plan that includes Cowork, and a Google account for the Drive connector or the documented local-folder fallback.
A short readiness check and setup guide are sent before the cohort begins, and the team supports you through setup, so you arrive ready to build.
Bring at least one repetitive, non-sensitive work task you would like to improve, or use Ratio Africa’s fictional practice materials.
Mac or Windows laptop
•Stable internet connection
•Paid Claude plan with Cowork
•Latest Claude Desktop application
•Google account or provided local-folder alternative
•One non-sensitive work task to improve
01Do I need to know how to code already?
Foundations starts assuming you have never opened a terminal. You will learn how to use an AI Agent to achieve what years of Engineering experience would accomplished. Intermediate and Advanced both require more intense terminal usage and you will taught how to go about it.
02What tools do I need?
A laptop running macOS, Windows, or Linux. A paid Claude account. No paid tools beyond a Claude subscription is required.
03What if I miss a live session?
Every live session is recorded and shared within 24 hours. Missing one is normal. Missing two makes catching up hard because the cohort builds fast. Instructor support is there for catch-up questions.
04Why is AI Agents Foundations priced so low?
The first Foundations cohorts are priced low intentionally. The early seats subsidise alumni building, honest feedback, and the case studies the next cohorts are sold against. The standard price rises by Cohort 3.
05What is the difference between AI Agents and AI Software?
AI Agents teaches you to build autonomous systems that scout, draft, push, and monitor on their own, usually on a schedule. AI Software teaches you to build apps and sites with embedded LLM features that users interact with directly. Most buyers belong in AI Agents.
06Can my company pay or invoice?
Yes. Email connect@ratioafrica.com with your company details for a proforma invoice. For teams of five or more, the Private and corporate route is usually a better fit.
07What if I need to cancel?
Plans change. If you need to cancel, the full terms are set out in our Refund Policy. Email connect@ratioafrica.com and we will help.
Price
Registration.
Open for registration
Ready to register?
Register for this cohort
Ready to register?
August 10 to 22, 2026
Registration is open. Share your details and Ratio Africa will send you to secure checkout.