Systems Software Engineer - ML Inference & GPU Clusters

raydar

San Francisco (CA)

In loco

USD 200.000 - 300.000

Tempo pieno

4 giorni fa

Candidati tra i primi

Generatore di candidature

Una candidatura fatta su misura per questo lavoro — un curriculum e una lettera di presentazione personalizzati, perfettamente in linea con l'annuncio.

Supera i filtri ATS

Vantaggi offerti da questo lavoro

Health coverage

Dental coverage

Vision coverage

Unlimited PTO

Parental leave

Descrizione del lavoro

Raydar is seeking a Systems Software Engineer to work on low-level performance and infrastructure for a production ML serving platform in San Francisco, CA. You will tune software for speed, run large compute clusters, and interact with customers in a high-autonomy, high-ownership team.

Responsibilities include delivering fast support for new models, improving the serving stack with batching and quantization, writing GPU kernels, and operating diverse compute clusters in a cost-efficient,

Competenze

Bachelor's degree in Computer Science or equivalent.
Strong background in backend/infrastructure or systems engineering.
Excellent communication with customers and cross-functional teams.

Mansioni

Deliver fast support for newly released open-source models, tuned for low latency and high throughput.
Improve the model serving stack with batching, caching, decoding strategies and quantization.
Write and tune low-level GPU kernels across several hardware vendors.
Design, deploy and operate compute clusters built from mixed hardware.
Run production inference at scale with reliability and cost efficiency; engage with customers.