Completed30 September 2026, 18:30–19:30 BST60 minutes, including questions
Choosing and pricing an AI tool stack
Session details
Getting reliable output from an AI assistant
How to frame a task, what context to supply, and how to check what comes back before you act on it. Covers the categories of work these tools handle badly.
Status
Completed
When
7 October 2026
20:30–21:30 BST
Length
60 minutes, including questions
Where
Online, live
Cost
Free
Ask about the next series
This session has finished. Email us to hear when the next series is announced.
Outcomes
What you leave with
- A way of framing a task that stops the model guessing at what you meant
- A check you can run on any output in under a minute, before you act on it
- The categories of work these tools are bad at, so you stop trying
- Your own reusable prompts for the two jobs you do most often
Who it is for
Anyone using AI daily who suspects they are only scratching at it.
What to bring
A real task from your own week that you would like to hand over.
Format
Running order
No technical background needed. Questions are taken throughout rather than held to the end.
01
Treating the model as a contractor
Briefing a capable contractor who has never seen your business changes what goes in the message. We start there.
02
Supplying context
What to paste in, what to leave out and why more context is not reliably better.
03
Verifying output
A short checking habit that catches the failures that matter, without becoming a second job.
04
Writing your own prompts
You write two in the session, on your own work, and test them before you leave.
7 October 2026, 20:30–21:30 BST
Getting reliable output from an AI assistant
Ask about the next series
This session has finished. The next session is listed on the schedule.