Home | Prices | Contact Us | Courses: Power BI - Excel - Python - SQL - Generative AI - Visualising Data - Analysing Data
Introduction to Generative AI Course
Generative AI assistants, such as large language models (LLMs) have practical use and can increase our productivity today: they can engage in a conversation with us and plan, brainstorm, summarise and explain. AI assistants are also useful for helping with Excel challenges, for data analysis and for coding in computer languages such as SQL or Python
Who Should Attend
This course is for people who want to take advantage of Generative AI in their daily tasks. It is aimed at a general audience.
Pre-requisites
None. This is a beginner level course. No previous knowledge or experience of Generative AI is required.
Course Length
This is a full day course.
Content
The course covers:
- an introduction to Generative AI and an explanation of LLMs;
- get started with generative AI; sign up to a free service
- a tour of some current Generative AI services and LLMs: Microsoft’s Copilot, OpenAI’s ChatGPT, Anthropic’s Claude and Google’s Gemini;
- practical exercises on how to use LLMs effectively (such as prompt engineering) to be more productive in everyday tasks;
Exercises
Starter Exercise: Get familiar with some AI assistants
This is a group exercise. Attendees form four teams and each team has different specific tasks and a rather glib name.
- Brainstormers generate novel ideas and possibilities and think creatively
- Explainers describe how things work, explain the root cause of a problem and suggest how it can be fixed.
- Planners make a plan for a task or set of activities. This could be a tiny plan, that take a few minutes to execute, or a grand plan that is implemented over decades.
- Summarisers make short work of long documents.
- Drafters note the key points and the AI generates a polished, grammatically correct version.
Write better prompts
There is a short individual exercise to practise writing better prompts: those that encourage more accurate and relevant responses from the AI assistant. We may use the RICE Framework to help with this.
Advanced Exercise: Explore the full capabilities of the AI assistant
This is a group exercise. Attendees form teams and each team has different specific tasks and a rather glib name.
- Thinkers use the powerful reasoning, or chain of thought, versions of the AI assistant to attempt more challenging and larger tasks. They also can approach tasks in ways that are perhaps not obvious at first sight. The capabilities of AI assistants in generating, reviewing, improving and summarising text are well known. This team explores capabilities that are not so well known: reasoning about the physical world, doing algebra, writing a poem, and writing and speaking in languages other than English.
- Visualisers explore the AI assistant’s capability to both generate and understand images: for example, to describe an image, to read text in the image even if on signage, to understand and interpret particular types of images such as maps, business charts, and to perceive the emotional state of a person from their image.
- Notebookers provide the AI assistant with a set of documents and ask it to answer questions based solely on the contents of the document (and not to use its general knowledge and also not search the web). These documents could be a PDF or Word document, or a links to web pages.
- Customisers configure the AI assistant to respond in a certain way, by telling the AI assistant something about themselves and their interests and how they would like the AI assistant to respond to them, and giving the AI assistant specific “system” instructions. If using ChatGPT, they also explore custom GPTs, customised version of ChatGPT created by Open AI, the makers of ChatGPT, and other third parties, and specialised for a particular task such as a personal tutor or an image generator. If time allows , they may build their own custom GPT.
- Data Analysts import some dataset and the AI assistant will act as a data analyst: describe the data, clean the data if necessary, create some charts and provide some insight.
Course Style
The course has has both tutorials and group exercises. In tutorials, attendees follow the instructor step by step or prompt-by-prompt. In the group exercises, attendees work in groups of about 4 - 5 people in breakout rooms for about 25-30 minutes. Once back in the main room, a spokesperson for each group summarises their conclusions to the class.