liveBook · Manning

GPT builder

As explained in AI for Everyday IT

GPT Builder is an OpenAI interface for designing and configuring custom GPTs through conversational instructions and settings.

GPT Builder is a conversational OpenAI interface for creating custom GPTs. It turns a workplace task into a focused assistant by combining a role, instructions, conversation starters, knowledge files, and selected capabilities. section 8.3.1

A practical configuration workflow includes:

  • Describe the assistant’s purpose and let Builder generate an initial design. section 8.3.1
  • Use Configure to edit the name, description, instructions, and conversation starters. section 8.3.1
  • Add knowledge files and enable only capabilities such as web browsing, image generation, or code execution that the task needs. section 8.3.1
  • Save the GPT, then choose whether to share it by link or through the GPT Store. section 8.3.1

Ticket Analyst

Configure a GPT to analyze IT service tickets, categorize problems and resolutions, inspect comments and closure notes, and summarize trends or anomalies. Keep instructions specific so responses stay focused. Verify access settings carefully: code execution can create risks, and knowledge documents should not contain anything unsafe to download. section 8.3.1

Table 8.2 Ticket Analyst settings
Setting
Value
Name
Ticket Analyst
Description
Analyze IT service tickets to help categorize problems and resolutions.
Instructions
Your primary role is to analyze IT service tickets from spreadsheets, categorizing problems into major categories and subcategories. When ticket details are unclear, you’re expected to make educated guesses based on available information. You will delve into comments, work notes, and closure notes to understand problem resolutions. Besides categorizing individual tickets, you should also provide summaries of overall trends and patterns in the data, highlighting common problems, effective resolutions, and any notable anomalies. This comprehensive analysis will aid in improving IT service management by identifying areas for process enhancement and training initiatives.
Conversation Starters
1. Analyze this service ticket for me.
2. What category does this problem belong to?
3. Summarize the resolution of this ticket.
4. Identify patterns in these ticket closures.
Knowledge
We provided a file named task.csv which contains several examples of IT tickets which we can use for the custom GPT to analyze.
Capabilities
Web Browsing, DALL-E Image Generation, Code Interpreter
Actions
Advanced programming required, not used in this example, but allows the GPT to interact with external resources

Use the Ticket Analyst settings to inspect a complete name, description, instruction set, and conversation-starter configuration. section 8.3.1

How AI for Everyday IT covers it

How deep it goes
Explains setup settings and configuration choices, but does not cover model internals or detailed validation procedures.
Examples
Uses a Ticket Analyst configuration for categorizing service tickets and summarizing problems, resolutions, trends, and anomalies.
What you can do after
Build it from scratch by configuring a task-specific GPT and sharing the saved assistant.
How it connects
Connects custom GPTs with instructions, conversation starters, knowledge files, web browsing, image generation, and code execution.

Related concepts

This explainer was generated by Manning's knowledge graph from the text of AI for Everyday IT. The section links point to the parts of the book it draws on.