Please, let Stack AV know you found this job
on RemoteYeah.
This helps us grow π±.
Description:
Stack is developing revolutionary AI and advanced autonomous systems to enhance safety, reliability, and efficiency in modern operations.
The ML Training Team focuses on training models quickly, ensuring 100% GPU utilization, and scaling from 8 to 256 GPUs.
Responsibilities include setting up efficiency monitoring for training jobs, benchmarking and profiling customer jobs, creating standardized APIs for training datasets and evaluation metrics, optimizing dataloaders and training data formats, and optimizing distributed training configurations.
Requirements:
Candidates must have 5+ years of experience as a Software Engineer, ideally in building infrastructure or customer-facing products, with experience in autonomous vehicles or robotics being a plus.
Ideal skills include experience with ML platforms and applications, building scalable and reliable infrastructure, and a deep understanding of design tradeoffs.
Experience with model training, optimization, large data processing pipelines, and pushing GPU limits from Python to CUDA is preferred.
Candidates should have shipped ML products at scale and be proficient in high-performance C++.
Desired attributes include high customer empathy and the ability to communicate effectively with customers, as well as comfort in reading and keeping up with state-of-the-art ML literature.
Benefits:
Stack is committed to building a diverse and inclusive workplace, believing that diverse teams produce the best ideas and outcomes.
The company offers a culture of inclusion, entrepreneurship, and innovation across various dimensions such as gender, race, age, sexual orientation, religion, disability, and identity.
Candidates should be aware that employment may be contingent upon verification of residence, U.S. person status, and/or citizenship status due to compliance with U.S. national security laws and regulations.
Apply now
Please, let Stack AV know you found this job
on RemoteYeah
.
This helps us grow π±.