Responsibilities

• Ensure efficient implementation of models & systems with a focus on large-scale training.

• Identify and implement optimization techniques for massively parallel and distributed systems, including the underlying communication layer.

• Identify and remedy efficiency bottlenecks (memory, speed, utilization, communication) by profiling and implementing high-performance PyTorch code, deferring to Triton, CUDA, and lower levels as necessary.

• Work closely together with the rest of the research team to ensure systems are planned to be as efficient as possible from start to finish.

• Conduct research & experiments on state-of-the-art large-scale generative AI models with the goal to improve latency & throughput for training and inference.

Must have experience

• Experience training large models using Python & Pytorch, including practical experience working with the full development pipeline from data processing, preparation & dataloading to training and inference.

• Experience profiling GPU & CPU code in Pytorch for optimal device utilization (examples: torch profiler, NVIDIA Nsight systems/compute, memory profilers, trace viewers, custom tooling).

• Experience writing & improving highly parallel & distributed Pytorch code of large generative models, with familiarity in FSDP, Tensor Parallel, Sequence/Context Parallel, Pipeline Parallel etc.

• Experience working with transformer models and attention implementations.

  Good to have experience

• Experience with high-performance Triton/CUDA and writing custom PyTorch kernels and ops. Top candidates will be able to write fused kernels for common hot paths, understand when to make use of lower level features like tensor cores or warp intrinsics, and will understand where these tools can be most impactful.

• Experience writing high-performance parallel C++. Bonus if done within an ML context with Pytorch, like for data loading, data processing, inference code.

• Experience building inference / demo prototype code (incl. Gradio, Docker etc.).

At Luma AI, we believe that multimodality is critical for intelligence. To go beyond language models and build more aware, capable and useful systems, the next step function change will come from vision. So, we are working on training and scaling up multimodal foundation models for systems that can see and understand, show and explain, and eventually interact with our world to effect change.

We will deploy these systems to make a new kind of intelligent creative partner that can imagine with us. Free and away from the pressure of being creative. It's for all of us whose imaginations have been constrained, who've had to channel vivid dreams through broken words, hoping others will see what we see in our mind's eye. A partner that can help us show — not just tell.

Dream Machine is an early step to building that. Try it here

Why you should join us:

• Luma is bringing together the best team in the world to achieve our goal, from researchers to engineers and designers to growth operators

• Luma is not just a lab - we are deeply product focused and our vision merging AI models and delightful products is unique in the industry

• We build. We ship. Our early products have been wildly successful