pythonsqlstatisticsfraud detectionmachine learningpytorchcomputer visionnlpscikit learntrust and safety
Вакансия из Hirify Global, списка международных tech-компаний Для мэтча и отклика нужен Plus
Пожаловаться
Зарплата и рынок
развернуть
По международному рынку данных пока нет
На СНГ рынке: $3.6к/мес ($2.6к - $4.4к)
88
Хорошая вакансия
развернуть
У роли четкий цикл работы с ML-моделями до продакшена, а компания предлагает заметную поддержку обучения. Это сильная и хорошо описанная возможность в продуктовой компании.
Актуальный стек
Понятные задачи
Бюджет на развитие
Крупный продукт
Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Подключить Plus
Описание вакансии
Текст:
Hirify AI/Оригинал
TL;DR
Data Scientist (Machine Learning, Trust & Safety) (Machine Learning/Trust & Safety): Building and improving machine learning models that detect harmful content and suspicious behaviour across a large marketplace with an accent on noisy data, class imbalance, model evaluation, and production deployment. Focus on developing multimodal detection solutions, integrating reliable training and inference pipelines, and balancing detection coverage with false positives and member experience.
Location: Kaunas or Vilnius, Lithuania; hybrid work with 2 recommended office days per week
Gross monthly salary: €3,433–€4,650
Company
hirify.global is a Lithuania-founded second-hand marketplace connecting millions of members across more than 20 markets, with additional logistics and payments businesses.
What you will do
Develop, evaluate, and improve machine learning models detecting harmful content and suspicious behaviour at scale.
Engineer features and prepare training and evaluation datasets, addressing noisy labels, class imbalance, and data leakage.
Collaborate with engineers to integrate models into production and build reliable training and inference pipelines.
Define success metrics and evaluate improvements through offline analysis and online experimentation with Decision Scientists and Product Managers.
Monitor deployed models, investigate changes in effectiveness, and maintain models as behaviour and data evolve.
Write reliable, maintainable, and reproducible code, document work, share learnings, and contribute to the team roadmap.
Requirements
Practical experience developing machine learning solutions with a solid foundation in machine learning, statistics, and probability theory.
Proficiency in Python and SQL, with experience using ML libraries such as scikit-learn or PyTorch.
Hands-on experience preparing training datasets, engineering features, training models, and evaluating them against baselines and metrics.
Experience applying machine learning to tabular or behavioural data, natural language processing, or computer vision.
Familiarity with the production ML lifecycle, including deployment, monitoring, and retraining.
Ability to break down problems, work independently within an agreed scope, and seek input when needed.
Ability to communicate approaches, findings, and trade-offs clearly and collaborate with technical and non-technical colleagues.
Motivation to apply machine learning to member safety, with care for data privacy and the consequences of incorrect predictions.
Nice to have
Experience with fraud detection, content classification, or other Trust & Safety applications.
Experience with transformers, multimodal models, or LLM-based workflows.
Culture & Benefits
Hybrid work with 2 recommended office days per week; remote workation is available for up to 90 days per year in the EU, including 21 days globally, or 21 days worldwide for non-EU citizens.
25 working days of holiday, private health insurance, pension contributions, an Employee Assistance Program, and share options.
Home-office equipment and a budget of up to €540 for workplace furniture.
Annual learning budget starting at €3,000 and up to 10 working days over two years for development.
Team-building events, a monthly hirify.global shopping budget, and office-specific meals or lunch allowances.
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 →