machine learning engineer for advertising technology

Enfint; Berlin; Hybrid; EUR 85.000 - 135.000; Vollzeit

Benefits dieser Stelle

Healthcare

Mental health services

Paid holidays

Family leave

Equity options

Performance-based rewards

Work abroad up to 30 days

Zusammenfassung

Smartly — платформа на базе ИИ для рекламы, объединяющая медиа, креатив и аналитику, поддерживает крупные кампании и достигать бизнес‑результатов брендов и рекламодателей.

Ищем ML‑инженера для разработки и развёртывания ПО, работы со сборами данных и усиления MLOps и data‑платформы. Гибридный режим: 3 дня в офисе, возможна удалённая работа.

Qualifikationen

• 5+ лет опыта разработки и развёртывания производственного ПО.
• 2+ года опыта внедрения сервисов на базе ML.
• 2+ года опыта работы в облаке (AWS или GCP).
• Свободное владение Python.
• Практика ML-фреймворков PyTorch или TensorFlow.
• Знание MLflow или Kubeflow.
• Опыт построения масштабируемых конвейеров данных.
• Опыт работы над сервисами под SLA.
• Сильная аналитика и умение решать задачи.
• Коммуникационные навыки на английском.

Aufgaben

• Разрабатывать и разворачивать ML‑ынфраструктуру и сервисы.
• Работать с данными: изображения, видео, аудио и текст.
• Улучшать MLOps и data platform.
• Упрощать внедрение AI/ML приложений и их надёжность.
• Учавствовать в планировании спринтов и ретроспективах.
• Сотрудничать с продуктом, инжинирингом и инфраструктурой.

Kenntnisse

Python

ML frameworks

English communication

Analytics

Software development

Ausbildung

M.Sc. in a relevant field

Tools

MLflow

Kubeflow

AWS

GCP

PyTorch

TensorFlow

Jobbeschreibung

Описание

Smartly is an AI-powered advertising technology company transforming ad experiences for brands and consumers. Its advertising platform integrates media, creative, and intelligence to support large-scale advertising campaigns and deliver business outcomes for brands and advertisers.

Задачи

• Build ML-based software systems for advertising experiences and campaigns
• Work with datasets covering images, video, audio, text, and structured performance data
• Contribute to and strengthen Smartly’s MLOps and data platform
• Help productionize AI/ML applications more robustly and efficiently
• Contribute to team processes, work planning, retrospectives, and continuous improvement
• Improve the daily work of staff at hundreds of advertisers using Smartly
• Work with product, engineering, and infrastructure stakeholders to develop viable solutions for customer needs
• Share knowledge, run productive meetings, and participate in pair programming and debugging
• Keep up to date with ML innovations in generative AI, computer vision, natural language processing, and explainability.

Требования

• 5+ Years of experience developing and deploying production-quality software
• 2+ Years of experience delivering software services powered by machine learning
• 2+ Years of experience working with cloud infrastructure such as AWS or GCP
• Fluent in Python
• Hands-on experience with modern ML frameworks such as PyTorch or TensorFlow
• Knowledge of MLOps concepts and tools such as MLflow or Kubeflow
• Experience in feature engineering, model evaluation, diagnostics, and monitoring
• Strong foundation in linear algebra, statistics, and calculus
• Experience building scalable data pipelines for ML-based data processing
• Experience developing and maintaining services under SLA
• Analytical mindset with a problem-solving approach
• Passion for solving customer challenges quickly while making thoughtful architectural decisions
• Adaptability and resilience when facing new or ambiguous challenges
• Strong written and verbal communication skills in English
• Ability and willingness to work in a hybrid capacity from the office 3 days a week
• Nice to have: C++ or Java experience, M.Sc. in a relevant field.

Условия

• Healthcare packages and mental health services
• Paid holidays and family leave
• Equity options
• Performance-based rewards
• Competitive compensation
• Career development opportunities
• Option to work abroad for up to 30 days annually.