Feature Store Architect (f/m/d)

adjoe is redefining the future of mobile ads. Powered by advanced AI first-party data and world-class engineers weve perfected the offerwall experience for monetizing and scaling app publishers with solutions like Playtime and Arcade now the fastest growing rewarded advertising channel globally driving incremental engagement retention and revenue. At the same time were breaking into programmatic advertising with the intelligent bidding system of adjoe Ads. Together this ecosystem connects app developers to over 600 million users worldwide for scalable growth. Backed by a $100 million investment from Bertelsmann and operating from offices in Hamburg Boston Singapore and Tokyo adjoe is defining the next stages of app and ad experience right now. Join us.

Meet Your Team: Programmatic Solutions

The Programmatic Solutions team powers adjoes real-time ad network solution integrated with leading mobile mediation platforms on both Android and iOS.

Our focus lies in building adjoes own Ad Network a smart bidder leveraging machine learning to optimize campaign performance by predicting conversion probability and lifetime value in real time.

The team is responsible for both the adjoe Ads platform that decides which ad to display and our Android/iOS SDKs that render these ads and provide tracking events to the backend.
Weve also developed tools through which app publishers can manage their integration analyze monetization performance and assess ad user experience through dashboards and APIs.

For potential teammates across all disciplines working in the Programmatic Solutions team means applying your problem-solving skills in adjoes advanced programmatic ecosystem where milliseconds directly impact both revenue and relevance.

What You Will Do

• Build & Maintain a Feature Store: Design implement and optimize a scalable efficient and easily accessible feature store for machine learning applications.
• Develop Scalable Data Pipelines: Build and maintain robust data pipelines that process and transform large-scale event data daily and in near real time within our data lake and feature platform.
• Enable High-Impact Features: Extract clean and engineer features from raw data to enhance the relevance and accuracy of recommendations in our adtech platform.
• Ensure Data Quality Observability & Integrity: Implement data validation monitoring and governance processes to maintain accuracy consistency and reliability across all datasets.
• Bridge Data Science & Backend Teams: Act as the key link between data science and backend engineering ensuring seamless data integration and usage across the organization.
• Work in an International Environment: Join an international English-speaking team focused on scaling our adtech platform to new heights.

Who You Are

• You have 5 years of software development experience working on modern data engineering data stack.
• You have proven experience in working with Feature Stores ideally building one from scratch.
• You have worked extensively with real-time data streaming systems such as Kafka Kinesis or Pub/Sub.
• You have experience with systems handling several TB of data per day and multiple thousand events per second.
• You know how to identify bottlenecks in data pipelines and you have experience in scaling them up.
• You have strong Python/Golang/Java or any other high-level programming language skills.
• You have worked closely with Data Scientists on ML systems for live/online predictions.
• You know scheduling frameworks such as Airflow / Kubeflow.
• You know the concepts of data quality and how to apply them in production.
• You are familiar with relational and NoSQL databases.
• You know common concepts such as caching rate limiting or concurrency and you know when and how to apply them.
• Plus: You have (hands-on) experience in working with AWS Terraform and Kubernetes.

Our Tech Stack

• We strongly believe in self-hosted open source technologies backed by battle-proven AWS technologies.
• Apache Kafka as our event streaming system
• Go as our primary language for backend
• Kubernetes and Terraform to manage our infrastructure
• S3 and Druid for long-term data storage
• DynamoDB and Redis for short-term data storage and real-time aggregations
• TensorFlow and TensorFlow Serving for machine learning
• Prometheus Grafana the ELK stack and OpenObserve for logging and monitoring
• Airflow for orchestration
• and were always open to trying new technologies that suit each case best.

Scale at a glance

• Thousands of auctions handled every second
• 99% of the auctions handled within 150ms
• 100k ML predictions per second
• 2TB data ingested in real-time every day
• more than 50 Airflow jobs and other data pipelines

Fuel for the Journey: Benefits to Support Your Ambitions

• Invest in Your Future: Regular feedback and our development program support your growth helping you expand your skill set and achieve your career goals.
• Easy Arrival to adjoe: From signing to settling in Hamburg weve got you covered. Need a visa No problem. Ready to build your new life and career at adjoe in Hamburg We support every ambitionfrom learning German to a relocation bonus that helps you settle in and make Hamburg feel like home.
• Live Your Best Life at Work and Beyond: We work in a hybrid setup with 3 core office days plus flexible working hours. Enjoy 30 vacation days 3 weeks of remote work per year and free access to an in-house gym with lots of different fitness classes and mental health support through our Employee Assistance Program (EAP).
• Thrive Where You Work: Enjoy the Alster lake view from our central office with top-notch equipment fun open spaces and a large variety of snacks and drinks.
• Join the Community! Participate in regular team and company events including hackathons and social gatherings. We work together and we celebrate together too.