Quality Analyst, Not Quality Assurance.
Page 14
Review AI-driven test-prioritization approaches and assess the team's data, decision factors, and potential benefits and risks.
Summarize practical frustrations with AI testing tools: data privacy, functional limits, learning effort, and difficulty accessing tools.
Discuss AI-assisted bug reporting and the author's privacy and trial-access concerns, which prevented a full evaluation of the tools.
Assess AI for accessibility testing by reviewing ChatGPT-4's tool suggestions and weighing potential benefits and limits against project needs.
A 2024 trial of Postman's Postbot assistant, covering generated request tests, response visualization, field extraction, and API documentation.
Review a poll on short-term AI use in testing, the author's own adoption, and the opportunities and concerns raised by integrating AI into daily work.
Try Postman's AI assistant on API requests: generate response tests, create visualizations, save response fields, and draft documentation.
Review the team's QA workflow and where AI could improve test design and execution, then outline how to promote AI-assisted testing.
Day 12 reflects on trusting AI in testing, including risks such as inaccurate results, bias, and privacy, and how safeguards shape appropriate use.
Generate Auth0 registration test data for email and social sign-in with ChatGPT 3.5 and 4, then assess how well it fits each scenario.
Compare ChatGPT 3.5 and 4 test cases for online shopping tax scenarios, reviewing functional coverage and how prompt details shape the results.
Day 9 asks participants to evaluate a prompt and its output, then improve it using criteria such as clarity, specificity, relevance, and usefulness.
