13 дней назад
Member of Technical Staff (Platform)
Вакансия из Hirify Global, списка международных tech-компанийДля мэтча и отклика нужен Plus
Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Описание вакансии
Текст:
TL;DR
Member of Technical Staff (Platform) (Kubernetes/AI): Operating model inference and training workloads across cloud and customer-hosted Kubernetes with an accent on reliability, performance, and cost efficiency. Focus on building Kubernetes controllers, optimizing GPU-backed inference and data-heavy Python pipelines, and solving deterministic scheduling, recovery, isolation, and profiling challenges.
Location: Remote from São Paulo, Brazil
Company
builds a model platform for governed inference, training, post-training, and fine-tuning across cloud and customer-hosted Kubernetes environments.
What you will do
- Evolve Sophos, the Kubernetes-based online and batch inference runtime.
- Run large-scale batch inference on ephemeral jobs using multi-dimensional CPU, memory, and GPU admission control through Kueue.
- Build and extend the Sophos controller and Kubernetes custom resources.
- Optimize model inference engines and feature-processing pipelines using vectorized and columnar operations.
- Own training, post-training, and fine-tuning job execution across ’s cloud and customer dataplanes.
- Drive autoscaling, GPU serving, telemetry, performance, reliability, and cost optimization.
Requirements
- Experience running model serving or large-scale batch compute on Kubernetes.
- Experience building Kubernetes controllers or operators.
- Strong Python profiling and optimization skills for data-heavy pipelines.
- Ability to write production-quality code, participate in reviews, and operate systems in production.
- Strong focus on compute efficiency, serving availability, inference latency, batch throughput, and job completion reliability.
Nice to have
- Production experience with Ray, Ray Serve, or KubeRay.
- Experience with Kueue or other batch scheduling and admission-control systems.
- GPU serving and performance optimization experience.
- Experience with Arrow, Parquet, Lance, or other columnar data formats.
- Experience shipping software to customer-hosted Kubernetes, including GCP/AWS, GKE/EKS, or regulated financial-services environments.
Culture & Benefits
- Technical ICs operate as Members of Technical Staff with ownership of systems and outcomes.
- Work remotely from São Paulo.
- Success is measured through 99.9% serving availability, p95/p99 online inference latency, batch throughput, cost per prediction, GPU utilization, and reliable job completion.
Будьте осторожны: если работодатель просит войти в их систему, используя iCloud/Google, прислать код/пароль, запустить код/ПО, не делайте этого - это мошенники. Обязательно жмите "Пожаловаться" или пишите в поддержку. Подробнее в гайде →