В Roblox чётко очерченная senior-роль в ML-инфраструктуре с современными системами большого масштаба, но в обязанности также входят поддержка production и дежурства.
?Дежурства в команде
Современный ML-стек
Чёткая зона ответственности
Масштабный продукт
Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Подключить Plus
Описание вакансии
Текст:
Hirify AI/Оригинал
TL;DR
Senior Software Engineer (Machine Learning) - Economy ML (PyTorch/distributed systems): Building training infrastructure, backend services, and batch and real-time feature pipelines for ranking, recommendation, and pricing models across Roblox’s virtual economy with an accent on GPU optimization, low-latency serving, and reliable production deployment. Focus on scaling ML systems, managing model rollouts and experimentation, and balancing model quality, latency, freshness, and cost.
Location: San Mateo, CA, United States. Hybrid office schedule: onsite Tuesday, Wednesday, and Thursday, with optional presence on Monday and Friday.
Annual base salary: $243,290–$295,250 USD.
Company
Roblox builds a platform and tools for creating and experiencing 3D immersive digital experiences, including virtual economy products for users, creators, and developers.
What you will do
Build and operate distributed training infrastructure, data-loading systems, and reproducible pipelines for ranking, recommendation, and pricing models.
Co-design and optimize model training and inference, including GPU utilization, memory, throughput, latency, batching, mixed precision, and feature processing.
Design backend services and APIs that integrate model predictions into Marketplace, Payments, and Developer Monetization with latency budgets and graceful fallbacks.
Build batch and real-time data and feature pipelines focused on freshness, correctness, and training/serving consistency.
Own model packaging, deployment, versioning, shadow and canary rollouts, and A/B experiment integration.
Improve reliability and observability, contribute to the ML infrastructure roadmap, collaborate with platform and backend teams, and mentor engineers.
Requirements
5+ years of software engineering experience, including 2+ years building and operating ML infrastructure or ML-backed production services at high scale.
Strong Python skills and proficiency in at least one other backend language, with experience designing and operating distributed, low-latency services.
Production experience training and serving deep learning models with PyTorch, including multi-GPU and multi-node training and GPU inference deployment.
Experience profiling and optimizing utilization, memory, latency, and cost.
Experience building batch and streaming data pipelines and feature stores, such as Spark, Flink, or Kafka, with attention to data quality and training/serving skew.
Ability to operate production systems, including monitoring, alerting, debugging, and incident response.
Ability to collaborate with ML engineers, platform teams, product managers, and data scientists across Marketplace, Payments, and Developer Monetization.
Understanding of how latency, freshness, and cost affect model quality and product metrics.
BS, MS, or Ph.D. in Computer Science, Engineering, or a related technical field, or equivalent practical experience.
Nice to have
Experience with recommendation, search, or ads systems.
Culture & Benefits
Full-time employees are eligible for equity compensation and benefits.
Work includes collaboration with central ML platform, data infrastructure, Marketplace, Payments, and Developer Monetization teams.
Equal employment opportunities and reasonable accommodations are provided during the recruiting process.
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 →