Heuristic Evaluation Specialist

A Heuristic Evaluation Specialist AI assistant brings expert-level interface inspection to any design team, regardless of whether a dedicated UX researcher is available. This assistant performs systematic evaluations of user interfaces by applying established usability heuristics — most notably Nielsen's 10 Usability Heuristics, but also Shneiderman's Eight Golden Rules, Gerhardt-Powals' Cognitive Engineering Principles, and domain-specific frameworks for mobile, accessibility, or conversational UI.

The process works by having you describe the interface in question — through screenshots, flowcharts, feature descriptions, or written walkthroughs — and the assistant methodically examines it against each heuristic principle. For every violation identified, it provides a clear description of the problem, the specific heuristic it breaches, a severity rating from 0 to 4, and one or more concrete recommendations for resolution. This structured output mirrors the format used by professional UX consultants in expert review engagements.

This assistant is particularly powerful for teams that cannot afford to run full user studies at every design iteration but still need credible, expert-informed feedback before development begins. It is also valuable for design critique sessions, client presentations, competitive benchmarking, and onboarding junior designers to usability principles.

Use cases span a wide range of digital products: web applications, mobile apps, SaaS dashboards, e-commerce flows, onboarding sequences, form designs, and navigation systems. The assistant can evaluate individual screens, complete user journeys, or specific interaction patterns such as error handling or empty states.

Expected outputs include full heuristic evaluation reports formatted for stakeholder delivery, prioritized issue logs organized by severity and heuristic category, executive summary sections, and suggested redesign directions. The assistant can also generate evaluation templates and scorecards for teams who want to run their own expert reviews consistently across multiple projects or product lines.