Зона ответственности за production-сервис инференса описана четко, а задачи сфокусированы и актуальны; главный нюанс - дежурства.
?Дежурства
Современный стек
Инфраструктура для AI
Понятная зона ответственности
Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Подключить Plus
Описание вакансии
Текст:
Hirify AI/Оригинал
TL;DR
Software Engineer, Inference (AI): Building and operating production model-serving services for a high-performance AI cloud with an accent on observability, reliability, and inference performance. Focus on optimizing latency and throughput through micro-batching, concurrency tuning, caching, and GPU inference techniques.
Location: Sunnyvale, CA or Bellevue, WA, United States
Base salary: $120,000–$160,000 per year, plus discretionary bonus, equity awards, and benefits subject to eligibility.
Company
CoreWeave provides high-performance cloud infrastructure for GPU workloads across AI/ML, VFX, rendering, and real-time inference.
What you will do
Own a model-serving service end to end, including design, implementation, testing, deployment, and operations.
Build dashboards, alerts, traces, structured observability, and automated runbook steps.
Improve inference latency and throughput through micro-batching, concurrency tuning, and caching.
Participate in on-call rotations and lead post-incident action items for the service.
Review pull requests and provide design feedback in team reviews.
Requirements
0–3 years of software engineering experience, or equivalent experience building cloud or distributed services in production.
Proficiency in Python or Go, with familiarity with gRPC/HTTP APIs and containerized deployments.
Working knowledge of Kubernetes, Slurm, or Ray and CI/CD pipelines.
Experience with metrics and alerting using Prometheus, Grafana, or OpenTelemetry, plus structured logging.
Basic understanding of GPU inference concepts, including KV caching, mixed precision, and streaming tokens.
Must satisfy U.S. export-control access requirements as a U.S. person, an individual eligible to access the information without required authorization, or an individual eligible and reasonably likely to obtain the required authorization.
Nice to have
Experience with inference frameworks such as Triton, vLLM, TensorRT-LLM, Ray Serve, or TorchServe.
Hands-on performance profiling or benchmarking with tools such as wrk, vegeta, perf, or Nsight.
Contributions to reliability practices, including SLI/SLOs or postmortem improvements.
Culture & Benefits
Remote-friendly culture with a focus on curiosity, ownership, innovation, and disruptive technical work.
100% employer-paid medical, dental, and vision insurance.
401(k) with a generous employer match, life and disability insurance, and flexible PTO.
Catered lunch at office and data center locations and a casual work environment.
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 →