Synthesia is on a mission to make video easy for everyone. Born in an AI lab, our AI video communications platform simplifies the entire video production process, making it easy for everyone, regardless of skill level, to create, collaborate, and share high-quality videos.
We are looking for a ML Performance Engineer to join our team of 40+ Researchers and Engineers within the R&D Department. As a ML Performance Engineer, you will contribute to the design and development of high-performance solutions and own one or more projects for computationally optimizing large-scale model training and inference pipelines.
Your responsibilities will include:
* Evaluating, profiling, and optimizing compute resource usage for cost and time efficiency at training and inference times.
* Developing customized efficient solutions for inference pipelines and introducing or enhancing tooling for achieving optimal computational performance.
* Driving the adoption of best practices for large-model training, including checkpointing, gradient accumulation, and memory optimization.
* Introducing or enhancing tooling for distributed training, performance monitoring, and logging.
* Designing and implementing techniques for model parallelism, data parallelism, and mixed-precision training.
* Keeping updated on the latest research in model compression and advanced optimization methods.
To be successful in this role, you should be a ML engineer passionate about high-performance computing, with a background in Computer Science/Engineering and 3+ years of industry experience. You should have experience optimizing large models, developing CUDA/Triton kernels, and working with DL compilers.
We offer attractive compensation, private health insurance, hybrid work setting, and opportunities for career growth. If you are interested in working for a company that is impacting businesses at a rapid pace across the globe, please apply.
#J-18808-Ljbffr