Troiana AI Hub courses

Practical AI skill,
one course at a time.

Choose the AI platform you use. Each course gives you a clear sequence of lessons, practical exercises, and techniques you can use in real work.

Choose your course

Every platform. The same high standard.

Start where you work today, then come back for the next tool.

Prompt Library

300 practical prompts for real work.

Curated ChatGPT and Codex prompts, each with the right context, outcome, and use case—not a wall of generic templates.

Product100 Design100 Development100 Research100 Marketing100 Content Writing100
ProductChatGPT

Write a clear product requirements doc

Turn a rough product idea into a usable PRD with decisions and acceptance criteria.

Use prompt
DevelopmentCodex

Debug a failing production path

Create a disciplined investigation plan before changing code under pressure.

Use prompt
Content WritingChatGPT

Write an SEO content brief

Turn a search opportunity into a writer-ready brief.

Use prompt
Built for product, design, development, research, marketing, and content work. Browse all 600 prompts
Keep learning

Practice, then put it to work.

Test your knowledge Browse Prompt Library
How to use AI tools well

Stop collecting prompts, start building judgement

Most advice on how to use AI tools is a pile of prompts to copy. That approach breaks the moment your task differs from the example. What actually transfers is judgement: knowing which tool fits the job, how to frame a request so the model has what it needs, and how to check the result before you trust it. This hub is built around that idea. Instead of a prompt dump, it is organised the way you actually decide — pick the platform, pick what you are trying to do, then drill into a specific, tested knowledge card. The goal is that you leave able to reason about a new task, not just repeat an old one. New to the technology itself? Start with our foundational guide, What Are Large Language Models (LLMs)?, then come back and pick a platform.

Platform, then intent, then card

The four assistants covered here — Claude, ChatGPT, Cursor, and Gemini — overlap, but they are not interchangeable, and the fastest way to a good result is choosing the right one for the task. From there, every platform is broken down by intent: coding, writing, research, workflows, and the mistakes to avoid. Under each intent sit knowledge cards — small, self-contained techniques you can apply immediately. That structure mirrors how a decision actually flows: what am I using, what am I doing, and what is the specific move that works. You do not read the hub front to back; you navigate to the exact answer.

The habits that beat any single prompt

A few principles hold across every tool. Give context, a clear goal, and the format you want, rather than a one-line request. Match the model to the task — a fast model for quick work, a reasoning model when being right matters. Curate what the model sees rather than dumping everything at it; managing the context window well is often the single biggest lever on quality. And treat a fluent answer as unverified until you have checked it, because polish is not accuracy. These habits, not a magic prompt, are what separate people who get real value from AI from people who get plausible noise.

From single prompts to real workflows

The last step is to stop treating AI as a place you ask questions and start treating it as a component in how you work. A prompt you run repeatedly becomes a saved assistant; a complex task becomes a chain of focused steps; a capable model becomes something wired into your real tools and data. That is where the productivity gains compound. Pick a platform below to begin, or jump straight to the intent that matches what you are trying to do — every path leads to concrete, tested techniques rather than theory.

Frequently asked
Which AI tool should I use — ChatGPT, Claude, Gemini, or Cursor?

It depends on the task. ChatGPT is the broadest generalist with the deepest feature set. Claude leads for coding discipline, long-form writing, and careful reasoning. Gemini stands out for very long context and Google Workspace integration. Cursor is an AI code editor for working inside a real codebase. Many people use more than one and route each job to the tool that fits it best — this hub helps you decide.

How do I get better results from any AI tool?

Give context, a clear goal, and the format you want, instead of a one-line request; match the model to the task; curate what the model sees rather than overloading it; and verify anything that matters instead of trusting a fluent answer. Those habits transfer across every tool and matter more than any single prompt.

Do I need to learn prompt engineering?

You need the fundamentals — clear context, a defined outcome, and an example of what good looks like — but not a bag of tricks. The higher-leverage skills are choosing the right tool and model, managing context, and verifying output. This hub teaches those as decisions you can reason about, not scripts to memorise.

Is this hub for beginners or advanced users?

Both. Each platform starts with basics — accounts, first prompts, choosing a model — and goes up through workflows and advanced techniques like context engineering and tool use. The knowledge cards are tagged by level, so you can start where you are and go deeper as you need to.

Are these AI techniques free to use?

The techniques are free to read and apply here. The tools themselves each have a free tier that is enough to learn on, plus paid plans that add higher limits and stronger models. Most of what this hub teaches works on the free tiers; the paid plans matter once you rely on the tools daily.

How is this different from a list of ChatGPT prompts?

A prompt list gives you fixed recipes that break when your task changes. This hub is organised by decision — platform, then intent, then a specific technique — so you build judgement that transfers to new tasks. The aim is understanding why a technique works, not just a phrase to paste.