Data engineer for RTB

Wirtualna Polska Media; Warszawa; Hybrid; PLN 180,000 - 270,000; Full time

Benefits offered by this job

Hybrid work in WP

WP Academy training

Free access to WP products

心理咨询服务

Job summary

Wirtualna Polska Media ищет инженера по данным для построения и оптимизации пайплайнов, подающих данные в ML-модели, и обеспечения высокой скорости обработки в реальном времени. Вы будете работать с масштабируемой инфраструктурой и аналитическими базами, участвуя в разработке наблюдаемости и мониторов качества данных.

Требуется опыт работы с Python, SQL, Spark/Polars и колоночными базами. Ожидается работа в гибридном режиме: 3 дня в офисе и 2 дня удалённо в Варшаве, участие в WP Academy и

Qualifications

• 3+ лет коммерческого опыта в Data Engineering или смежной роли, связанной с обработкой данных.
• Опыт работы с Python и SQL; знание Spark и Polars.
• Навыки проектирования структур данных для продвинутой аналитики и систем ближе к реальному времени.

Responsibilities

• Разработка и оптимизация процессов подачи данных в модели.
• Проектирование и оптимизация слоя хранения данных, индексов и схем.
• Создание слоя наблюдаемости данных — дашборды и оповещения.
• Разработка инструментов мониторинга моделей и выявления дрейфа распределения.
• Реализация CI/CD практик для развёртывания предиктивных моделей.

Skills

Python

SQL

Spark

Polars

Columnar databases

Data processing

Go (Golang)

Tools

ClickHouse

StarRocks

Kubernetes

GitOps (ArgoCD)

Job description

Описание:

We are building the "brain" of an RTB bidder - a system that buys advertising space in real-time auctions on behalf of e-commerce clients. Our ML models decide which campaign to display, what the ad should look like, and whether and how much to bid. We operate at a large scale: hundreds of thousands of requests per second, tens of billions of ML model decisions per day, and hundreds of millions of auction log rows entering the database every day. As a Data Engineer, you will develop processes that deliver data to these models.

Задачи:

Building and optimizing processes that feed data to models. Designing and optimizing the data storage layer, including schemas and indexing for analytics performance and Real-Time systems. Creating an observability layer for input data, including dashboards and alerts that detect anomalies and declines in the quality of data feeding models. Developing tools for monitoring production models, including detecting distribution shift. Implementing CI/CD practices that support the process of creating, testing, and deploying predictive models.

Требования:

At least 3 years of commercial experience in Data Engineering or a related role involving data processing. Very good knowledge of Python and SQL. Experience processing large volumes of data using tools such as Spark and Polars. Experience with columnar analytical databases. Strong technical skills: programming skills supported by an understanding of how complex systems work. Experience working in a sector that operates with large-scale data. Experience designing and modeling data structures for advanced analytics and systems operating in near real time.

Будет плюсом:

• Knowledge of specific analytical engines: ClickHouse and/or StarRocks;
• Knowledge of Kubernetes and GitOps tools (ArgoCD);
• Experience in the RTB (Real-Time Bidding) or programmatic advertising industry;
• Experience in machine learning (ML) projects;
• Knowledge of Go (Golang)

Условия:

Work for a technology leader and first-choice partner - we create content, services, and entertainment that accompany millions of Poles every day. Free access to our products, such as Pilot WP and Audioteka - we want you to feel part of what we create. Worksmile and a flexible benefits package - you choose what matters to you. Access to free and anonymous psychological consultations. Possibility of hybrid work - we work 3 days in the office and 2 days from home, combining the freedom of remote work with the best of the office. Events that bring us closer - from seasonal activities (Children's Day is our favorite!) to webinars on health and neurodiversity. Opportunity to develop through training as part of the WP Academy program. A chance to take on new challenges and explore new directions within WP Holding companies - internal recruitment opportunities are within reach.