Rendering Systems Engineer - Simulation & Synthetic Data

World Labs•San Francisco, CA

21h•$250,000 - $325,000

About The Position

World Labs is a frontier AI research and product company advancing spatial intelligence, the next frontier beyond large language models. Co-founded by Dr. Fei-Fei Li, Justin Johnson and Ben Mildenhall, the company is pioneering world models that perceive, generate, reason, and interact with virtual and physical worlds. The company’s flagship product, Marble, transforms text, images, and video into fully navigable 3D worlds, unlocking applications across gaming, film, architecture, robotics, and immersive digital experiences. Backed by leading investors and with over $1B raised, World Labs is assembling a world-class team at the intersection of AI research and real-world deployment. We are seeking a Senior/Staff Rendering Systems Engineer to build and scale high-throughput, Unreal Engine–based rendering systems for synthetic data generation. You will own the low-level rendering and GPU performance work required to run many-world rendering efficiently across cloud GPU fleets. You will profile end-to-end workloads, diagnose performance/IO bottlenecks, and implement production-quality C++ and CUDA systems within and around Unreal Engine. You will also partner with infrastructure and data teams to scale distributed execution and integrate the renderer with internal simulation and data pipelines.

Requirements

Mastery in low-level Unreal rendering architecture and GPU programming.
Broad experience on distributed multi-GPU compute.
Core graphics performance engineering.
Strong background in performance engineering, including telemetry profiling, roofline analysis, latency/throughput optimization, and systematic root-cause analysis.
Deep understanding of Unreal rendering stack, familiar with its codebase at platform-agnostic rendering layer, the command lists abstract layer, and the RHI abstraction layer, alongside practical experience customizing engine source at various layers and shader pipelines. Familiar with its GPU resource lifetime and synchronization, as well as async compute.
Working knowledge of Nanite, Lumen, and associated graphics debugging suites.
Advanced low-level GPU optimization skills (CUDA/Vulkan), with emphasis on kernel-level tuning, memory hierarchy management, and memory bandwidth efficiency.
High proficiency in C++, CUDA, Python (and possibly Rust), accompanied by practical Vulkan implementation experience.
Proficiency in rendering material adjustment and custom shader development pipelines.

Nice To Haves

3-4+ years of engineering experience within AI research labs, machine learning enterprises, or robotics and autonomous vehicle organizations.
Familiarity with distributed data interconnects and collective communications primitives (e.g., NVLink, NCCL).
Proven track record deploying or serving large-scale generative, diffusion, spatial, or video foundation models.
Hands-on creation of performance-profiling, telemetry, and observability toolchains for real-time graphics pipelines and GPU workloads.

Responsibilities

Own the low-level rendering and GPU performance work required to run many-world rendering efficiently across cloud GPU fleets.
Profile end-to-end workloads, diagnose performance/IO bottlenecks, and implement production-quality C++ and CUDA systems within and around Unreal Engine.
Partner with infrastructure and data teams to scale distributed execution and integrate the renderer with internal simulation and data pipelines.
Deep experience modifying Unreal Engine’s renderer to support run independent worlds or cameras across GPUs and build the cloud job-orchestration layer.
Maximize hardware throughput and GPU efficiency via custom parallelization, caching strategies, and high-performance interconnect communication.
Integrate rendering system seamlessly with proprietary internal toolchains and data pipelines.
Perform rigorous profiling, roofline analysis, and root-cause diagnosis to resolve compute and I/O limits.
Collaborate with research teams to accelerate experimental iteration and enhance system robustness at scale.