Job description

ABOUT WORLD LABS World Labs worldlabs.ai 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 profiles.stanford.edu, Justin Johnson linkedin.com and Ben Mildenhall linkedin.com, the company is pioneering world models that perceive, generate, reason, and interact with virtual and physical worlds. The company’s flagship product, Marble marble.worldlabs.ai, 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. ROLE OVERVIEW We are looking for a Performance Engineer to make World Labs’ models train and serve as fast as the hardware allows. Running large generative world models at scale is a novel systems problem. You will find the bottlenecks — in kernels, in the serving path, in the training loop, in how we use our GPUs — and eliminate them. Your ownership is technical and concrete: the throughput you unlock, the latency you cut, the utilization you win back, and the correctness you hold while doing it. You will work up and down the stack, from low-level tensor and kernel optimization to fleet-wide serving efficiency, in close partnership with the researchers whose models you are accelerating. This is a hands-on, individual-contributor role. You will profile, design, build, and ship code directly. WHAT YOU WILL DO - Optimize inference and serving end to end — latency, throughput, batching, caching, and scheduling — to serve our models efficiently at production scale. - Write and tune GPU kernels (CUDA, Triton) for hot paths; drive kernel fusion, memory- and bandwidth-bound optimization, and low-precision (FP8/INT8) execution. - Optimize training throughput and GPU utilization: parallelism strategies, communication/compute overlap, mixed precision, and eliminating pipeline stalls. - Build performance models, profiling workflows, and observability that make throughput, latency, cost, utilization, and their tradeoffs legible across the stack. - Own numerical correctness across precision, kernel, and hardware changes — treating correctness as part of performance, not separate from it. - Partner with researchers to productionize models for serving and to make experiments run faster and more reliably. - Where needed, work on the distributed systems that training and inference run on — but the core of the job is squeezing the most out of every GPU. KEY QUALIFICATIONS You should excel at the fundamentals below — we index on inference, serving, GPU optimization, and training performance. Distributed-systems breadth is welcome, but secondary. - Strong performance-engineering foundations: profiling, roofline analysis, latency/throughput optimization, and disciplined root-cause investigation. - Deep GPU programming and optimization experience (CUDA and/or Triton) — kernel-level tuning, memory hierarchy, and bandwidth optimization at scale. - Hands-on experience optimizing inference and serving for large models: batching, KV/prompt caching, quantization, and low-latency, high-throughput sampling. - Hands-on experience optimizing training performance: parallelism, distributed communication, mixed/low precision, and utilization. - Working knowledge of ML framework internals (PyTorch and/or JAX; torch.compile, XLA, or similar compiler paths). - Strong proficiency in Python, with the ability to drop into C++/CUDA (and Rust or Go) as the work demands. - High-ownership mindset — you measure yourself by throughput shipped and latency cut, not tickets closed. PREFERRED QUALIFICATIONS - Experience at an AI lab or ML-native company, optimizing systems used directly by researchers and productionizing research code. - Low-precision and numerics depth: FP8/INT8 quantization, mixed-precision, and detecting numerical regressions across hardware platforms. - Distributed systems for large-scale training and inference — collective communication (NCCL), interconnects (NVLink), model and tensor parallelism, and fault tolerance. A strong plus, but not a substitute for the core skills above. - Experience serving generative, diffusion, video, or 3D/spatial models — not just text LLMs. - Multi-accelerator experience (GPU plus TPU or Trainium) and partnering with hardware vendors on accelerator capabilities. - Building performance-modeling and observability frameworks for GPU utilization and cost. WHO YOU ARE - Fearless Innovator: We need people who thrive on challenges and aren't afraid to tackle the impossible. - Resilient Builder: Impacting Large World Models isn't a sprint; it's a marathon with hurdles. We're looking for builders who can weather the storms of groundbreaking research and come out stronger. - Mission-Driven Mindset: Everything we do is in service of creating the best spatially intelligent AI systems, and using them to empower people. - Collaborative Spirit: We're building something bigger than any one person. We need team players who can harness the power of collective intelligence. We're hiring the brightest minds from around the globe to bring diverse perspectives to our cutting-edge work. If you're ready to work on technology that will reshape how machines perceive and interact with the world - then World Labs is your launchpad. Join us, and let's make history together. Logistics: Location: This role is based in San Francisco, California; we work in-person in the office 5 days a week. Visa sponsorship: We sponsor visas. While we can't guarantee success for every candidate or role, if you're the right fit, we're committed to working through the visa process together. Benefits: We offer generous health, dental, vision and 401K benefits, flexible PTO, paid parental leave, and relocation support as needed. Equal Employment Opportunity World Labs is an equal opportunity employer. We do not discriminate on the basis of race, color, religion, sex, sexual orientation, gender identity, national origin, age, disability, genetic information, veteran status, or any other characteristic protected under applicable law. We welcome all qualified applicants and are committed to providing reasonable accommodations throughout the hiring process upon request.

Requirements

Must-haves

We are looking for a Performance Engineer to make World Labs’ models train and serve as fast as the hardware allows.
This is a hands-on, individual-contributor role. You will profile, design, build, and ship code directly.
Optimize inference and serving end to end — latency, throughput, batching, caching, and scheduling — to serve our models efficiently at production scale.
Write and tune GPU kernels (CUDA, Triton) for hot paths; drive kernel fusion, memory- and bandwidth-bound optimization, and low-precision (FP8/INT8) execution.
Optimize training throughput and GPU utilization: parallelism strategies, communication/compute overlap, mixed precision, and eliminating pipeline stalls.
Build performance models, profiling workflows, and observability that make throughput, latency, cost, utilization, and their tradeoffs legible across the stack.
Own numerical correctness across precision, kernel, and hardware changes — treating correctness as part of performance, not separate from it.
Partner with researchers to productionize models for serving and to make experiments run faster and more reliably.
Where needed, work on the distributed systems that training and inference run on — but the core of the job is squeezing the most out of every GPU.
You should excel at the fundamentals below — we index on inference, serving, GPU optimization, and training performance. Distributed-systems breadth is welcome, but secondary.
Strong performance-engineering foundations: profiling, roofline analysis, latency/throughput optimization, and disciplined root-cause investigation.
Deep GPU programming and optimization experience (CUDA and/or Triton) — kernel-level tuning, memory hierarchy, and bandwidth optimization at scale.
Hands-on experience optimizing inference and serving for large models: batching, KV/prompt caching, quantization, and low-latency, high-throughput sampling.
Hands-on experience optimizing training performance: parallelism, distributed communication, mixed/low precision, and utilization.
Working knowledge of ML framework internals (PyTorch and/or JAX; torch.compile, XLA, or similar compiler paths).
Strong proficiency in Python, with the ability to drop into C++/CUDA (and Rust or Go) as the work demands.
High-ownership mindset — you measure yourself by throughput shipped and latency cut, not tickets closed.
Fearless Innovator: We need people who thrive on challenges and aren't afraid to tackle the impossible.
Resilient Builder: Impacting Large World Models isn't a sprint; it's a marathon with hurdles. We're looking for builders who can weather the storms of groundbreaking research and come out stronger.

Nice-to-haves

Experience at an AI lab or ML-native company, optimizing systems used directly by researchers and productionizing research code.
Low-precision and numerics depth: FP8/INT8 quantization, mixed-precision, and detecting numerical regressions across hardware platforms.
Experience serving generative, diffusion, video, or 3D/spatial models — not just text LLMs.
Multi-accelerator experience (GPU plus TPU or Trainium) and partnering with hardware vendors on accelerator capabilities.
Building performance-modeling and observability frameworks for GPU utilization and cost.

Application prep

Use the role’s language and application details while tailoring your materials.

Keywords to mirror

aimlpythonrustgo

Platform: ashby
Effort: quick

Benefits

Benefits detected in the listing.

Dental

Vision

401k

Pto

Parental Leave