AI Platform Engineer (Serverless Training and Inference) (Python)

The role is open in Poland, Austria, Lithuania, and other locations, with compensation varying by country; the work format is not specified.

Зарплата не указанаПольшаКипрВеликобританияГерманияИспанияПортугалияОАЭЛитваВенгрия
Текст:
Оригинал доступен только пользователям с подпиской

Location

The role is open in Poland, Austria, Lithuania, and other locations, with compensation varying by country; the work format is not specified.

About the company

A global fintech company with over 80 million customers and a team of more than 15,000 people working in offices and remotely. Its technology organisation builds the infrastructure and features behind a widely used financial app.

Responsibilities

  • Design and build self-service platforms for model training, distributed training, and inference on shared GPU infrastructure
  • Develop APIs, SDKs, and automated workflows for workload submission, infrastructure provisioning, and execution
  • Build orchestration and distributed execution using Flyte, Ray, or similar technologies, including scheduling, queueing, resource allocation, and autoscaling on Kubernetes-based GPU environments
  • Ensure reliable workload lifecycle management, monitoring, multi-tenant isolation, and cost visibility to improve platform reliability and GPU utilisation

Requirements

  • Strong Python engineering skills and a track record of designing, building, and operating scalable backend platforms in production
  • Experience building or extending ML platforms for training or inference on GPU infrastructure, with expertise in Kubernetes, containerised workloads, and cloud infrastructure
  • Familiarity with workflow orchestration and distributed execution frameworks such as Flyte, Ray, or equivalent technologies, plus understanding of distributed training, GPU allocation, scheduling, and autoscaling
  • Experience designing reliable distributed systems with fault tolerance, observability, and operational ownership, supported by a STEM degree or equivalent practical experience and solid computer science fundamentals

Похожие вакансии MLOps инженер

Сайты компаний

Risk Ops Engineer

Автоподнятие
Зарплата не указанаГрузияТбилиси
Текст:
Оригинал доступен только пользователям с подпиской

Location

Limassol, with the specific work format and any geographic restrictions not specified.

About the company

A global technology company operating in the lending and fintech space, building credit decisioning and processing systems for a large international user base.

Responsibilities

  • Own the full lifecycle of the decision-making system and credit processing engine, covering architecture, deployment, monitoring, code quality and performance tuning.
  • Lead a focused developer team, handling code reviews, release management and performance evaluation in an agile setup.
  • Work with risk analysts and data scientists to turn risk rules and ML scoring models into production software, and integrate credit bureaus, collection and fraud detection services via REST APIs and event-driven pipelines.
  • Modernize delivery workflows and introduce automated regression testing to shorten time-to-market for credit strategy changes and reduce reliance on legacy release cycles.

Requirements

  • 7+ years of backend engineering experience with deep fintech domain knowledge in credit decisioning, loan processing, debt collection or automated decisioning engines.
  • Familiarity with a popular decision engine such as gorules, n8n, camunda, SAS RTDM or FICO Blaze Advisor.
  • Strong Golang and Python skills, plus experience with relational and non-relational databases (MySQL, AuroraDB, DynamoDB, GBQ).
  • Hands-on with CI/CD tools (GitLab CI, GitHub Actions), Docker, monitoring tools (Prometheus, Grafana) and AWS; high-volume microservices experience is a plus.
Сайты компаний

Senior Software Engineer, Infrastructure Automation

Зарплата не указанаСШАКанадаТоронтоУдалёнка
Текст:
Оригинал доступен только пользователям с подпиской

Location

Toronto; remote-friendly within a globally distributed setup, with the US team working alongside European colleagues.

About the company

A rapidly scaling technology company operating a cloud data platform across more than 25 countries. Its cloud platform infrastructure group builds and maintains the foundational backend, internal tooling, and standards that support engineering teams and newly acquired products.

Responsibilities

  • Design, deploy, and maintain fault-tolerant distributed infrastructure across hundreds of clusters and thousands of nodes in multi-cloud, multi-region environments (AWS, GCP, Azure)
  • Build internal tooling, abstractions, and automated workflows that let other engineering teams self-serve infrastructure and reduce manual operations
  • Develop cloud-native platform services with GitOps and automation, lead mid- to long-term migrations, and drive security hardening
  • Improve observability and production debugging with partner teams, and take part in a working-hours on-call rotation of about one week every other month

Requirements

  • Relevant software engineering or platform infrastructure experience operating large-scale, stateful, fault-tolerant distributed systems, such as hundreds of clusters, dozens of cloud accounts, or thousands of nodes
  • Proficiency in Go and/or Python with Bash familiarity and a developer-first approach to building reliable services and automation
  • Deep hands-on experience with managed Kubernetes, major public clouds (strong preference for AWS and/or GCP), and Infrastructure-as-Code tools such as Terraform, Terragrunt, and Atlantis
  • Proven GitOps and CI/CD experience with tools like ArgoCD and Helm, plus strong ownership, autonomy, and production debugging skills in a globally distributed organization
hh.ru
ано цисм

ML-разработчик (NLP/LLM)

ано цисмАвтоподнятие
Зарплата не указанаРоссияМоскваУдалёнка

Центр изучения и сетевого мониторинга молодёжной среды – аккредитованная IT-компания, созданная по поручению Президента России в 2018 году.

Наша миссия – разработка социально значимых технологических решений для защиты детей и подростков от деструктивного контента в цифровом пространстве. Через создание комплексной мониторинговой системы мы формируем безопасную среду для нового поколения.

Мы ищем в команду ML-разработчика, ответственного за разработку NLP-моделей и работы с большими языковыми моделями (LLM). Мы работаем с колоссальным объёмом текстов, разрабатываем собственные NLP-модели и платформу использования LLM для автоматизации бизнес-процессов.

Стек технологий

  • PyTorch, HuggingFace Transformers
  • Python, NumPy, Pandas, Sci-kit Learn
  • Docker, Kubernetes
  • NVidia Triton, TensorRT, ONNX, vLLM
  • Jira, Confluence, Git

Чем предстоит заниматься

  • Разработка NLP-моделей для работы с данными социальных сетей и социальных медиа
  • Разработка LLM-навыков для автоматизации различных задач
  • Взаимодействие с отделом разметки данных и отделом DevOps
  • Внедрение моделей в промышленную эксплуатацию

Ждем от кандидата

  • Опыт работы от 3 лет
  • Опыт работы или реализации практических проектов в области машинного обучения
  • Глубокое понимание математических оснований машинного обучения и анализа данных
  • Проактивность, готовность вникать в задачу
  • Знание метрик оценки качества LLM-workflow (LLM-навыков)
  • Опыт работы с реляционными базами данными
  • Качественный чистый код

Что мы предлагаем

  • Чёткие процессы разработки, взаимодействия с проектными командами и DevOps-инженерами
  • Большой кластер GPU, включающий в том числе самые современные production-ready ускорители
  • Официальное трудоустройство по ТК РФ в аккредитованной IT-компании
  • Конкурентный уровень з/п, размер обсуждается по итогам собеседования и зависит от квалификации
  • Формат работы - офис/гибрид/удаленно
  • Гибкое начало рабочего дня, график 5/2, выходные: суббота, воскресенье
  • Непрерывное обучение и развитие профессиональных навыков

HireSeeker собирает вакансии со всех площадок и присылает только релевантные. Бесплатно.