About Lucky Robots

Lucky Robots is building the world’s first game engine for robotics — a real-time 3D simulation and training platform where AI learns to move, manipulate, and understand the physical world. We generate massive amounts of high-fidelity training data through large-scale simulation.

Our flagship product, Lucky Engine, is a AAA-grade 3D simulation environment built for one purpose: making large-scale robot training and testing faster, more accessible, and more accurate than anything on the market. Our bar for fidelity is uncompromising — if you plugged a simulated robot into a computer over USB, the output should be indistinguishable from the real thing.

We’re a small, fast-moving team across AI, simulation, physics, and graphics. We ship quickly and turn ideas into working software fast.

The Role

Own the UI and UX of Lucky Engine end-to-end. You’ll define how a serious 3D simulation engine looks, feels, and behaves — from the viewport interaction model down to every inspector, panel, and shortcut.

This is not a marketing site, mobile app, or admin dashboard. It is a desktop 3D engine — dense, visual, and powerful, but immediately legible. Software that someone can open for the first time and start working in, and still grow into for years.

The catch: our users are not all 3D experts. They include robotics researchers, ML engineers, hobbyists, startups, and enterprise teams who want to simulate robots and generate training data — many have never opened Blender or Unreal. Your job is to design a tool as powerful as a AAA game engine and approachable enough that someone who has only ever used Cursor or VS Code can be productive in it.

That tension is the design problem.

The existing tools in this category are notoriously painful to use. We are setting a new standard.

What You’ll Design

• Viewport & 3D interaction — camera, gizmos, selection, the “feel” of working in 3D space
• Scene graph and hierarchy — how users build and navigate complex robotic scenes
• Node-based editors — behaviors, materials, data flow, scripting
• Asset editors and content browser — mesh, material, texture, rig, and animation editors; the surfaces where users import, organize, and refine 3D content
• Robot and sensor configuration — joints, kinematics, cameras, lidar, IMU, contact sensors
• Simulation inspection & debugging — playback, scrubbing, state inspection, sensor visualizations
• Timeline and recording — capturing and reviewing training data runs
• Job orchestration — large-scale parallel simulation jobs and datasets
• Training data inspection — reviewing, filtering, and curating engine output
• High-density property panels and inspectors — modular, fast, scriptable
• AI Agent surface — a natural-language interface that lets users describe what they want the engine to do (e.g. “have the G1 robot walk to the table and pick up the cup”) and have the scene set itself up

You’ll work directly with the founders and engineers, set the design direction, and ship.

Design Vision

Lucky Engine should feel like the best of three worlds:

• The density and capability of a AAA game engine — Unreal Engine, Unity, Frostbite, Ubisoft’s Snowdrop, Decima. Modular panels, dockable layouts, deep inspectors, fluid workflows.
• The approachability of modern developer tools — Cursor, VS Code, Linear. Clean typography, sensible defaults, surfaces that don’t punish a first-time user.
• The information density of serious creative tools — Blender, Figma, Houdini — without the visual noise or steep onboarding cliff.

In short: extreme information density without clutter, contextually surfaced inspectors, a modern 2026 look, and an onboarding path that doesn’t require the user to already know a game engine.

Professional-grade software that anyone can pick up. A researcher with a PhD, a hobbyist tinkering on weekends, a startup founder prototyping a robot, and an enterprise team running thousands of simulations should all feel at home. Power without prerequisites.

Platforms

Lucky Engine ships on Windows and Linux as first-party platforms — both are equal citizens, not afterthoughts. Our users span both, often using each for different parts of their workflow, and the engine has to feel native everywhere it runs. Knowing the visual language, window management, font rendering, and file-dialog conventions of each platform matters — especially the parts of Linux desktop design that good cross-platform tools get right and bad ones don’t.

Who You Are

You care deeply about how complex software feels to use. You think in pixels, interactions, and information hierarchy, and you can hold the mental model of a complex tool in your head and find the elegant cut through it.

Examples of the kinds of work that resonate with us:

• 3D tools or game engines — Blender, Maya, Houdini, Unreal Engine, Frostbite, Unity, or anything in that space
• Developer tools or IDEs — Cursor, Visual Studio, VS Code, JetBrains IDEs
• Complex data, design, or creative tools — Figma, Notion, Ableton, DaVinci Resolve
• Smaller, niche, or indie software that’s exceptionally well-designed — internal tools, plugins, mods, side projects. If it’s dense, technical, and good, we want to see it.

You should be comfortable owning both UI and UX — information architecture, interaction, prototyping, visual design, and design systems. Strong opinions on dense UI, discoverability, and making complex tools feel approachable matter more to us than years on a resume.

Fluency with Windows and Linux desktop conventions is important. You’ve designed (or daily-driven) professional software on both, and you understand what makes a tool feel native rather than ported. The engine ships first-party on both platforms and has to feel at home on each.

Bonus

• 3D modeling, game development, or robotics experience as a user — not just a designer
• You can build interactive prototypes (Figma, code, or in-engine mockups)
• A point of view on what existing robotics simulation tools get wrong

Engagement

• Contract-to-hire. Project-based contract with flexible hours. Strong fit converts to a founding designer role with full-time terms and equity.
• Melbourne (hybrid). Regular in-person collaboration with the founders and engineering team.
• Compensation: competitive, based on experience.
• Direct collaboration with founders and engineers — no design-by-committee, no layers.

How to Apply

Send us (nur@luckyrobots.com):

1. Your portfolio — especially anything dense, technical, or desktop-shaped. Side projects, internal tools, and unshipped work all count.
2. A short note on a piece of software you think has great UX, one you think has terrible UX, and what you’d do about the latter.
3. Anything else you want us to see — writing, mods, plugins, GitHub, Dribbble, demos.

We read everything. Real work beats a polished resume.