Apply on Employer Site

MeshyAI · 2 weeks ago

Machine Learning System Engineer

San Francisco Bay Area

Full-time

Hybrid

Mid, Senior Level

MeshyAI is a leader in 3D generative AI, dedicated to unleashing creativity through innovative technology. The Machine Learning System Engineer will work on building an end-to-end ML framework for 3D, focusing on both training and inference challenges while collaborating with researchers to advance 3D and Spatial AI.

Computer Software

Responsibilities

Work closely with researchers to co-design the next frontier of 3D & Spatial AI

Build and debug on top of modern PyTorch, for maximum parallelism and efficiency, and build clean and intuitive training infrastructure for our in-house foundational models

Identifying bottlenecks and optimizing for high throughput & efficient distributed model training across hundreds to thousands of GPUs

Implementing and maintaining 3D specific custom operators in Triton or CUDA

Implementing and maintaining novel data-loading framework and libraries

Building efficient inference endpoints with complex multi-stage model pipelines

Optimizing models through compilation, fusion, quantization, etc

Qualification

Machine LearningPyTorchC++High Performance GraphicsDistributed TrainingData PipelinesPerformance Oriented MindsetEagerness to LearnCommunication Skills

Required

Experience in machine learning or high performance graphics

Solid practical understanding of at least one machine learning framework (e.g. PyTorch, JAX)

Strong ability to write beautiful and maintainable code in Python and/or C++

Ability to learn fast and dive into new concepts or complex codebases

Performance and efficiency oriented mindset, with a strong interest in the tiniest detail

Strong communication skills for working in a globally distributed team

Preferred

A strong passion to navigate through the PyTorch internals, with hands-on experience in areas like torch.compile, fully_shard (FSDP2) APIs

Experience with building Triton kernels

Experiences with large-scale distributed training, familiarity with modern parallelization techniques: DP, TP, CP, PP, zero redundancy optimizers, etc

Experience with diffusion models in 3D or video

Experience with low precision bf16 or fp8 training