Frontend Code Evaluator - Expert

Mercor connects elite creative and technical talent with leading AI research labs. Headquartered in San Francisco, our investors include Benchmark, General Catalyst, Peter Thiel, Adam D'Angelo, Larry Summers, and Jack Dorsey.

About the role

As a Frontend Engineer — Web Replication Preference Rater, you will evaluate the accuracy of web page replications by comparing a reference page with two candidate versions.

Responsibilities

• Render a reference page and two candidate replications side by side at desktop width.
• Judge which candidate is the closer reproduction by assessing layout fidelity, including box model, typography, color, image handling, and overflow.
• Inspect underlying markup with browser devtools to distinguish genuinely correct replications from those that merely appear correct.
• Evaluate responsive behavior and semantic quality, ensuring proper use of flexbox and grid.
• Write structured rationales for every judgment, ranking defects by importance in precise language.
• Flag ties, ambiguous cases, and broken task items rather than forcing a preference.

Requirements

• 3+ years of professional web development experience, primarily in frontend or full-stack work.
• Fluency in hand-written HTML and CSS.
• Working command of browser devtools.
• Enough JavaScript to read scripts and understand DOM manipulation.
• Comfort in a terminal for local server setup.
• Strong written English and discipline to justify judgments.

Qualifications

• Preferred: Prior RLHF, preference labeling, or model evaluation work.
• Preferred: Code review or technical assessment background.
• Preferred: Pixel-perfect design-to-code experience.
• Preferred: Web accessibility expertise (WCAG, ARIA).
• Preferred: Familiarity with LLMs' typical failures in codegen.
• Preferred: Web scraping or DOM parsing experience.

Pay

$90/hour