UX/UI Designer For AI Training (Figma, RLHF)

About OpenTrain

OpenTrain is the #1 platform for finding and building careers in AI training and data labeling. We connect designers, linguists, annotators, and other contributors with projects that let them build a durable freelance career teaching AI systems.

• Work remotely and build a unified AI training portfolio you control.
• Discover projects across the industry and apply in minutes; creating an account is free.

About AI Training Work

AI training (also called data labeling or human feedback work) is the human side of building modern AI: people prepare and evaluate examples that models learn from. Designers who contribute UX/UI examples help models understand good product decisions, accessibility, and real-world interaction patterns.

This role is part-time, flexible, and remote—ideal for practicing designers who want to shape how design-aware AI systems behave without prior AI experience.

• Contribute real-world design artifacts and written rationales that improve model reasoning.
• Projects often use evaluation ratings and RLHF-style feedback to refine generative systems.

The Role — Graphic Designer AI Training Expert

OpenTrain is recruiting UX/UI designers to supply high-quality design inputs and clear decision rationales used to train and evaluate AI systems. You will apply your professional UX/UI skills, primarily using Figma, to create examples and critique designs that teach models how to design, reason, and prioritize usability and accessibility.

• Employment type: Contractor, part-time.
• Time commitment: 20+ hours per week.
• Compensation: Paid hourly (PAY_PER_HOUR), $15–$30 USD/hour (typical listed rate $30/hr).
• Work format: Remote and asynchronous; English required.

What You'll Do

You will produce real design artifacts, explain your design decisions, and evaluate model outputs against project specifications. Clear, concise communication is as important as the visual examples you provide.

• Create UX/UI examples in Figma that follow project briefs and accessibility/usability best practices.
• Write short rationales that explain trade-offs, patterns, and accessibility considerations.
• Evaluate model-generated designs and rate them according to quality, clarity, and adherence to brief (evaluation rating / RLHF tasks).
• Maintain consistency and meet quality thresholds across tasks in a remote, asynchronous workflow.

Requirements

The role emphasizes professional design expertise and clear communication. No prior AI training experience is required; domain knowledge in UX/UI and a strong portfolio are essential.

• Proven UX/UI design expertise and strong portfolio of digital product projects.
• Proficiency with Figma and producing clear deliverables from it.
• Ability to articulate complex design concepts in written and verbal forms.
• Familiarity with usability, accessibility, and responsive design best practices.
• Skill in analyzing and critiquing designs with constructive, objective feedback.
• Ability to work autonomously and efficiently in a remote, asynchronous setting.

Helpful Background

Experience with AI projects or data annotation workflows is a plus but not required. Familiarity with evaluation or RLHF-style tasks will help you ramp up faster.

• Previous involvement in model evaluation, usability research, or design critique programs can be beneficial.
• Comfort working with short written rationales and structured rating rubrics.

How the Project Works & How To Apply

Work is delivered through the project's labeling/evaluation platform. You will complete tasks to spec, submit Figma assets and written rationales, and provide evaluation ratings when asked.

To apply, create or use your OpenTrain profile, include a strong Figma-based portfolio, and indicate availability for 20+ hours/week. OpenTrain and OpenTrain will use your profile to assess fit and next steps.

• Labeling type: IMAGE-focused inputs with evaluation rating and RLHF tasks.
• Language: English required. Worldwide applicants accepted.
• Company: OpenTrain (hiring via OpenTrain).