Post-Training ML Research Engineer - Scale Transformers

Baseten; San Francisco (CA); Presencial; USD 200.000 - 275.000; Jornada completa

Ventajas ofrecidas por este puesto de trabajo

Competitive compensation

100% medical coverage

Generous PTO policy

Paid parental leave

401(k)

Exposure to ML startups

Descripción de la vacante

A leading AI company in San Francisco seeks a research engineer to enhance post-training ML models. You will develop in-house tools, work with diverse model architectures, and interact with various technologies including Kubernetes and GPU computing. The role offers competitive compensation, comprehensive benefits including medical coverage and PTO, and a collaborative environment to drive innovation in AI.

Formación

• Deep understanding of ML model training techniques.
• Experience with GPU computations and performance profiling.
• Willingness to tackle complex problems collaboratively.

Responsabilidades

• Build in-house tooling for model training efficiency.
• Collaborate across technical stacks for systems-level concepts.
• Support customers’ post-trained ML models.

Conocimientos

Understanding of modern ML techniques

Advanced experience in PyTorch

Understanding of transformer training parallelism

Profiling distributed GPU programs

Ability to perform roofline analysis

Familiarity with HPC and distributed computing

Solid fundamentals in operating systems

Creativity and problem-solving skills

Herramientas

PyTorch

TensorFlow

Jax

Kubernetes

Slurm

Dask

Descripción del empleo

A leading AI company in San Francisco seeks a research engineer to enhance post-training ML models. You will develop in-house tools, work with diverse model architectures, and interact with various technologies including Kubernetes and GPU computing. The role offers competitive compensation, comprehensive benefits including medical coverage and PTO, and a collaborative environment to drive innovation in AI.