Skip to main content
Careers›Image-to-Code / Visual Reconstruction Engineer

Image-to-Code / Visual Reconstruction Engineer

RemoteGlobalFront-EndHTML/CSSReactSVG/Canvas

What this pays

Rates are set by the client project, not by us, and we do not know them until a contract is in place. Work is paid hourly or per task depending on the engagement. We would rather tell you that than publish a figure we cannot stand behind.

About the Role

Global | Contract | Mid-Level to Senior | 3+ years experience We are seeking experienced Front-End, UI, and Design Engineers for a project focused on recreating visual references through code with extremely high visual accuracy. Engineers receive an image or visual reference, recreate it using code, render the result, compare it to the original, and iterate until the implementation closely matches the source. Ideal candidates Front-End Engineers, UI Engineers, Design Engineers, Creative Developers, Visualization Engineers, Graphics Engineers, Web UI Engineers, and Interactive Developers. Not a strong fit Backend-only, DevOps, infrastructure, or primarily systems-focused engineers without significant visual or UI implementation experience. Apply if you enjoy pixel-perfect work, have a strong visual eye, and are comfortable repeatedly comparing, refining, and improving your code until the output closely matches a reference image. Location: global. Candidates may be located anywhere.

Requirements

  • •3+ years of software engineering experience
  • •Strong Front-End, UI, Design Engineering, Creative Development, Graphics, or Visualization background
  • •Excellent HTML/CSS and modern web development skills
  • •Strong understanding of layout, positioning, spacing, typography, color, and visual detail
  • •Exceptional attention to detail and the ability to spot small visual differences
  • •Strong coding and problem-solving skills
  • •Preferred: JavaScript/TypeScript, React, SVG/Canvas, Python, Pillow, NumPy, Matplotlib, SVGWrite, or CairoSVG

Responsibilities

  • •Translate images and visual references into code
  • •Recreate layouts, spacing, typography, colors, borders, shadows, icons, and other visual details
  • •Render your work and compare it against the source
  • •Identify subtle discrepancies and refine the implementation for pixel-level accuracy

Why Join?

  • Fully remote and open to candidates anywhere in the world
  • Contract work on a live AI training project
  • An objective definition of done: the render matches the reference
  • Project-based, on your own schedule

Quick Apply

Uses the profile you already have — nothing to re-enter.