CA0286 - UI/UX Designer

A thought experiment: What if every pixel you placed could either save a developer 40 hours of debugging hell or accidentally trigger a data pipeline that costs someone $50K? Welcome to designing for AI platforms where the stakes are existential and the users are unreasonably smart. We're CloudAngles Digital Transformation. We build AI-powered platforms where machine learning meets production reality—Testing Aide, Dangles, MLangles, Brew Content. Our users are platform engineers, data scientists, and ML practitioners who can smell bad UX from three screens away and will rage-quit your interface for a terminal if you waste their cognitive load. Here's the uncomfortable truth about this role: You'll design interfaces for workflows that don't exist in any design system. Multi-agent orchestration dashboards. Token usage visualization across 47 different LLMs. Test coverage heatmaps for synthetic data generators. This is uncharted territory—no medium articles to copy, no Dribbble shots to reference. You'll be making design decisions that become conventions. You'll also fail. A lot. Because designing for enterprise AI is designing for Murphy's Law at scale. That beautiful animation? Crashes on edge cases. That intuitive metaphor? Confusing to someone who thinks in DAGs and tensors. Your job is to fail faster, learn harder, and ship interfaces that make impossibly complex things feel obvious. You're clinically obsessed with: Information architecture that respects working memory limits (Miller's Law isn't optional) Interaction patterns that reduce cognitive friction, not just look modern The philosophical difference between "easy to use" and "easy to understand" What you've actually done: 2-3 years designing B2B SaaS, dev tools, or enterprise platforms where users have opinions and Slack receipts. Portfolio proving you can organize chaos into coherent systems. Fluent in Figma, conversational in frontend constraints, and comfortable debating component APIs with engineers who have thoughts.

You'll design interfaces for workflows that don't exist in any design system. Multi-agent orchestration dashboards. Token usage visualization across 47 different LLMs. Test coverage heatmaps for synthetic data generators. This is uncharted territory—no medium articles to copy, no Dribbble shots to reference. You'll be making design decisions that become conventions. You'll also fail. A lot. Because designing for enterprise AI is designing for Murphy's Law at scale. That beautiful animation? Crashes on edge cases. That intuitive metaphor? Confusing to someone who thinks in DAGs and tensors. Your job is to fail faster, learn harder, and ship interfaces that make impossibly complex things feel obvious.