pythonsqldbtrisk managementllmbigquerymachine learningexperimentationtrust and safetystatistical inference
Вакансия из Hirify Global, списка международных tech-компаний Для мэтча и отклика нужен Plus
Пожаловаться
Зарплата и рынок
развернуть
€107к-145к/год
€79к€80кмедиана€96к
По похожим ролям (17) ~ €80к/год
На СНГ рынке: $4.4к/мес ($3.7к - $5.9к)
89
Крутая вакансия
развернуть
Чётко очерченная роль уровня Staff в крупной продуктовой компании даёт широкое влияние на ML-системы в продакшене и сопровождается сильными возможностями для развития и хорошим пакетом льгот.
Актуальные технологии
Понятная зона ответственности
Бюджет на развитие
Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Подключить Plus
Описание вакансии
Текст:
Hirify AI/Оригинал
TL;DR
Staff Data Scientist (Trust & Safety) (Machine Learning/LLM): Designing large-scale detection, risk-scoring, and decision-making systems for protecting a marketplace community with an accent on statistical evaluation, layered inference, and continuous risk assessment. Focus on combining siloed models into a calibrated architecture, measuring detection coverage and harm, and setting technical direction across teams.
Location: Amsterdam, Netherlands; hybrid model with 2 recommended office days per week. Remote work is permitted for up to 90 days per year in the EU, including 21 days globally; for non-EU citizens, 21 days worldwide.
hirify.global operates a second-hand marketplace connecting millions of members across more than 20 markets, alongside shipping and payment businesses.
What you will do
Design end-to-end machine learning architecture for detection, risk scoring, routing, and decision-making.
Establish continuous risk assessment across registration, listing creation, and ongoing member activity.
Define end-to-end measures for detection effectiveness, including precision, coverage, missed harm, and computational cost.
Integrate team-level models into a coherent, calibrated detection system and close detection coverage gaps.
Set shared technical standards, prototype critical components, build reference implementations, and communicate trade-offs to senior technical and business leaders.
Requirements
8+ years of experience in Data Science, Machine Learning, or a similar role, including significant experience owning production models or model evaluation at scale.
Experience leading cross-team technical work without formal authority and setting standards adopted by other teams.
Deep expertise in statistical inference and experimentation, including A/B testing, precision and recall measurement under imperfect or biased labels, ground-truth sampling, and causal reasoning.
Understanding of the production ML lifecycle, including evaluation, threshold setting, retraining, drift, feature stores, and monitoring.
Hands-on experience evaluating LLM-based systems, or a strong interest in developing rigorous evaluation methods for them.
High proficiency in SQL and Python and familiarity with a modern data stack such as BigQuery and dbt.
Ability to turn cross-system problems into clear plans and explain trade-offs to technical and non-technical audiences, including senior leadership.
Fluent English, written and spoken, is required.
Nice to have
Experience in Trust & Safety, fraud prevention, content moderation, or risk management.
Culture & Benefits
Hybrid work with flexibility to decide when to work from home and when to meet in person.
25 days of paid annual leave, work-from-home allowance, home office equipment support, and up to €540 for workplace furniture.
Share options, pension plan, lunch benefit, hirify.global shopping budget, flexible benefits budget, and public transport support.
Learning budget starting at €3,000 annually and up to 10 working days over two years for professional development.
Employee Assistance Program, team-building events, dog-friendly office, and company bike leasing scheme.
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 →