Description
A chat window is the least useful way to use a language model. This half day is about the API underneath it: getting machine-readable output, giving the model tools, and keeping cost and latency under control in something you actually ship.
What you will leave with
- Call an LLM API directly and handle its failure modes
- Get reliable structured output instead of prose you have to parse
- Use function/tool calling to let a model trigger real actions
- Measure and control token cost, latency and context limits
Agenda
| The API beneath the chat box: messages, roles, parameters | |
| Structured output — schemas, validation and why prose parsing fails | |
| Tool / function calling: giving a model real capabilities | |
| Cost, latency and context: what actually drives your bill | |
| Failure modes — refusals, truncation, hallucinated fields | |
| Build: a small tool-using integration of your own |
Who it is for
Developers and technical staff who have used ChatGPT and now need to build something dependable on top of a model.
Prerequisites
Ability to read and write basic Python or JavaScript.
Format
- Duration: Half day (13:30–17:00)
- Level: Intermediate
- Group size: maximum 10 attendees
- Delivery: live online, or in person by arrangement
Need this for a whole team, or extended to full length on your own systems? We deliver every workshop as a bespoke on-site course too — talk to us.
Reviews
There are no reviews yet.