ML Engineer (Python)
Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Описание вакансии
TL;DR
ML Engineer (Python): Developing and deploying high-load ML services for a video hosting platform with an accent on model integration, inference optimization, and data pipeline management. Focus on building scalable production services using Python, gRPC, and modern infrastructure tools to ensure stability and performance for millions of users.
Location: Onsite in Cyprus (Relocation provided)
Salary: From €5,000 net per month
Company
An international IT holding with over 15 years of experience, specializing in high-load video hosting products.
What you will do
- Develop high-load ML services in Python using gRPC.
- Integrate ML models into production, including feature preparation and API development.
- Optimize model inference through batching, ONNX, and latency reduction.
- Build and maintain data pipelines using Airflow for daily model updates.
- Ensure service stability via monitoring, logging, and alerting.
- Collaborate with backend and DevOps teams to manage feature stores.
Requirements
- Strong Python skills, including asyncio, typing, and service architecture.
- Experience developing backend services.
- Basic understanding of ML models and experience integrating them into production.
- Proficiency with Docker containerization.
- Experience working with APIs and databases.
- Must be willing to relocate to Cyprus.
Nice to have
- Experience with ONNX, Triton Inference Server, Airflow, or ClickHouse.
Culture & Benefits
- Work on a high-load product with real-world impact.
- Flat organizational structure with minimal bureaucracy.
- Comprehensive benefits package including medical insurance and tax compensation.
- Professional training and language courses.
- Full relocation support to Cyprus, including visa assistance, tickets, and initial accommodation.
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 →