Start with a useful result
You do not need to begin a UX side project with an entire application. A better first project may be one clearer flow, a useful template, or a small review service. AI can help you explore directions and make something testable sooner, giving you more opportunity to practice the judgment behind the interface.
Start with a task you can observe
Choose a person trying to do something concrete, such as finding a suitable appointment and understanding what happens after booking. Walk through the current flow. Note the expectations, questions, and failure states before deciding that the answer is a new visual style.
Turn the idea into a bounded offer
A pilot might review one flow with up to five screens, document three friction points, and propose a small prototype. State what is included and what needs separate work. Do not describe a quick review as a comprehensive accessibility audit or a full research study.
Use AI to expand alternatives
Supply the task, evidence, constraints, and current flow. Ask for several distinct approaches and the tradeoffs behind them. You choose which directions are worth prototyping. Generated personas and imagined interviews can be exercises, but they must not appear in the evidence column.
Prototype the decision you need to test
Make the smallest interaction that reveals whether the idea is understandable. Include relevant errors, empty states, and recovery. AI coding or design help can speed production, but you still need to inspect keyboard use, focus, labels, content, and responsive behavior in the actual result.
Watch a relevant person try it
Give a realistic task without explaining where to click. Observe the attempt, ask neutral follow-up questions, and record limitations of the session. A small study can surface issues, but it does not establish a population-wide success rate. Keep participant information and consent handled appropriately.
Tell the story with evidence
Show the original task, your assumptions, the artifact, what you observed, and what changed. Explain your own decisions and how AI contributed. If you offer a paid pilot, agree on scope and rights to any public example. A strong learning record does not need invented conversion metrics.
A brief you can use today
Write five lines: the person I want to help; the task they struggle with; the smallest result I can deliver; the evidence I will collect; and the time and money I will put into the first test. Ask your AI tool to challenge what is unclear before it drafts a plan. Keep decisions and source material together so you do not have to reconstruct the project in every conversation.
Keep the tool stack small
Begin with a tool you can access and one place to organize the work. Add hosting, a database, a domain, or specialist creative tools only when the experiment requires them. Subscriptions are not evidence of progress. Your first useful output may be a document, a prototype, or a manual service that helps you learn what should be automated later.
Learn the process by building something small
In Two Mondays, Ashley will build a product and show the steps from validation and planning through design, coding, testing, and marketing. Bring a small project and practice the workflow. The workshop meets October 26 and November 2, 2026, from 7–9 p.m. Eastern, on Zoom with a West Palm Beach in-person option. The $99 ticket covers both Mondays and optional setup help and office hours. Paid software and hosting are separate; you can follow the demonstrations without buying every tool mentioned here.
Explore an AI for UX team workshop
See the deliverables, boundaries, and next step before you book.
Explore the scope ↗Keep exploring
A better-looking redesign is a hypothesis. Here is how to test it.
How to prioritize UX friction without giving your product a made-up score