MIKE OICHERMAN
PRODUCT DESIGN · RESEARCH · AI-ASSISTED PROTOTYPING & BUILD
LOOPQA
Back
An autonomous QA product that explores a live web app, writes its own tests, and returns bug reports complete with a recording, root cause analysis, and a suggested fix.
Role Product Designer Consultant
Timeline Q1 2026 - On-going
Company Replay.io
The Problem
The pace of AI-assisted development has outrun the pace of QA. As more developers (and non-developers) ship code written or scaffolded by LLMs, deployment frequency has gone up, and so has the surface area for bugs. Most teams don't have a QA hire, and most "vibe coders" have no QA process at all. Replay QA is built on Replay's deterministic browser-debugging engine, so it can explore an app like a real user, capture exactly what happened at runtime, and hand back a recording plus root cause instead of a vague bug ticket.
Goals
I was brought on as a consultant to research UX patterns and user behavior around a pre-built prototype/early version of the product, followed by a design system and a UX/UI rework.
OBJECTIVE 1 Identify product market fit & analyze user behavior and usage patterns.
OBJECTIVE 2 Create a design system & design 1st iteration based on the research.
Research & Findings
GoalsMap the creator's journey, identify user needs, pinpoint high-friction states, and uncover the "AHA" moment.
Process
~30 Live User Interviews Conducted with early Loop QA users, spanning solo vibe-coders, small teams embedding QA into AI coding platforms, and engineering teams running Playwright/Cypress to understand current QA habits and pain points.
Screenshare recordings reviewed Observed real user sessions to identify friction points and how people interpreted bug reports.
Competitive Analysis Investigated how similar products differentiate themselves and identified potential feature gaps in our offering.
Key Findings
With more developers writing code via LLMs, bugs and deployment frequency are both rising; the need for QA has never been more evident, yet fewer teams have dedicated QA resourcing to meet it.Automated QA per deployment would be of significant value, teams wanted quality gates that run continuously, without someone having to manually kick off a test pass.Most of the users had a clear expectation of what Builder should do, and have had experience with vibe coding tool, indicating somewhat of an oversaturation in the market.
Testing & Validation
Method50 x 60-minute video interviews with screen-sharing, where users interacted with the prototype, sharing their thoughts and feedback.
Key Findings & Iterations
Get to value as fast as possible. The AHA moment isn't "the tool ran", it's finding more than one bug and being handed a root cause plus a potential fix. Friction between dropping in a URL and that moment needed to be reduced as much as possible.
Less is more. Over-delivering information, verbose reports, redundant detail, every possible signal - increased cognitive load rather than trust. Trimming reports down to what's actionable mattered more than showing everything the system found.
Design System
Foundation I created and maintain the design system, built on the Shadcn framework and mapped to brand guidelines, giving the product a consistent visual language grounded in a well-supported, extensible component base.
Velocity Working with AI-ready components let design and engineering move quicker, components were built to be legible to both designers and AI coding tools, shortening the gap between design and shipped UI.
Coverage Beyond the base framework, I created a set of additional components to support UX patterns unique to Loop QA, covering interactions the underlying framework didn't account for out of the box.
Live Product View Product