machine learning engineer in automotive

CarOnSale; Berlin; On-site; EUR 90,000 - 130,000; Full time

Benefits offered by this job

Пенсионное обеспечение 20% от работод

Deutschlandticket

Субсидия на спорт

Виртуальные опционы на акции

Современная IT-камплект

Job summary

CarOnSale ищет ML-инженера для ведения моделей от handoff до продакшна, включая упаковку, развёртывание и мониторинг. Вы будете отвечать за путь подачи и инференса, включая артефакты конвейера и совместимость с feature-store.

Требуется 2+ года опыта, strong Python и опыт работы с облачными ML-платформами (SageMaker/Vertex AI/Databricks). Гибридный график в Германии, соцпакет и опционы на акции.

Qualifications

• 2+ года продакшн ML-инженерии с владением моделями после handoff
• Написание типизированного тестируемого Python-кода
• Опыт работы с управляемыми ML-платформами (SageMaker/Vertex AI/Databricks/Azure ML)

Responsibilities

• Владение моделями от передачи до продакшна: упаковка, развёртывание, мониторинг и решения по готовности
• Обеспечение надёжности продакшна через обнаружение дрейфа, мониторинг производительности и реагирование на инциденты
• Управление путём сервинга и вывода, артефактами конвейера и совместимостью с feature-store
• Ревью дизайна модели и методологии оценки перед релизом
• Выявление утечки данных и ошибок в процессе разработки
• Расширение общей платформы без внедрения проектной логики в общий код
• Установление инженерных стандартов по мере масштабирования платформы

Skills

ML инженерия

Python

Code review

AWS

Tools

SageMaker

Vertex AI

Databricks

Terraform

Job description

Описание

CarOnSale is an AI-powered platform for B2B used car trading in Europe. It connects software, pricing intelligence, logistics, and financing for more than 40,000 buyers across over 20 countries.

Задачи

• Own models from handoff through production, including packaging, deployment, monitoring, and readiness decisions;
• Maintain production reliability through drift detection, performance monitoring, alerting, and incident response;
• Own the serving and inference path, including fitted pipeline artifacts, inference entry points, monitoring hooks, and feature-store parity;
• Review model design and evaluation methodology before release;
• Identify data leakage, backward-window errors, and weak evaluation during development;
• Extend the shared platform while keeping project-specific logic out of shared code;
• Set engineering standards for the platform as it scales across the organisation.

Требования

• Have 2+ years of production machine learning engineering experience with ownership of models after handoff;
• Write strong, typed, tested, production-grade Python code;
• Review the work of others;
• Understand machine learning sufficiently to challenge problem framing, feature engineering, model selection, and evaluation methodology;
• Have hands-on experience with a managed ML platform such as SageMaker, Vertex AI, Databricks, or Azure ML;
• Have experience with feature stores, CI/CD for machine learning, AWS, and Terraform;
• Actively use tools such as Claude, ChatGPT, or Copilot in daily work;
• English at C1 level, written and spoken;
• German is not required;
• Nice to have: Snowflake and dbt, mentoring colleagues or reviewing their work, comfort operating where the answer is not defined yet;

Условия

• Hybrid schedule with 3 days in the office and 2 days remote;
• 25 “Work from Anywhere” days per year;
• 28 Days of annual leave;
• Two annual career and development conversations;
• Company pension with a 20% employer contribution;
• Fully paid Deutschlandticket;
• FitX membership or Urban Sports Club subsidy;
• Virtual stock options;
• Modern IT setup;
• Structured onboarding with a buddy programme and social events;
• Dog-friendly office.