You maintain and innovate on our big data platform. Our analytics platform is a central data source for our website and app on one side and analysts and data science teams on the receiving side. Y

How do you make our customers happy?

Working in the Platform Interaction Data team, you help bol to understand how our millions of customers interact with our e-commerce platform. What was a user’s customer journey? Where did they click? How did they get to a page? We have built a platform to help answer these questions!

What do you do as a PID (Platform Interaction Data) Machine Learning Engineer?

In this role, you maintain and innovate on our big data platform. Our analytics platform is a central data source for our website and app on one side and analysts and data science teams on the receiving side. You will be responsible for extracting meaningful information from terabytes of user traffic. We are looking for someone that can do actor classification, based on traffic… characteristics. This data is at the core of a lot of decisions, so it is a big responsibility to do this in an awesome way.
Due to the complexity, breadth and dependencies of the platform, the role is served with a bit of a learning curve. But we have your back.
During your first few weeks, we expect no more (and no less) of you than getting to know your colleagues, getting to know our way of working (hint: freedom within a framework) and getting up close & personal with bol’s technical environment.
You are part of a team of 3 Engineers with a focus initially on Bot identification and
classification. We are looking for someone that can build Machine Learning powered products and take ownership of the entire data science lifecycle.

What technologies do we use?

We have some components that still run in our Data Centre. The rest of the platform is running in the Google Cloud Platform (GCP)

As for the tooling, we use:
• Python
• Kotlin/Java
• Redis
• Docker and Kubernetes
• Vertex AI for Machine Learning
• BigQuery for Data Storage
• Jupyter notebooks for data analysis
• Gitlab (for version control and CICD)
• Apache Airflow for ETL
• Apache Flink as our streaming framework
• Apache Kafka as our streaming platform
Do not worry if you are not familiar with all the above – we will make sure you get up to speed. Most importantly, we are curious to what you can bring to improve our systems!

Current topics and challenges include:
• Bot identification and classification.
• Making our self-service data platform easier to use for analysts and Data Scientists.
• Make sure that all the human user interactions are measured.
• Analysis and investigation into production incidents involving bad actors and finding ways to support other product teams so mitigate their impact in the future.
3 reasons why this is (not) for you
• You are eager to help You want to deliver a platform towards the organization that is central to understanding our customers with amounts of data that most engineers would run from.
• You love simplicity Sure, our platform should adhere to guidelines and be compliant, but you always go for simplicity: you understand that developers should be able to deliver their solutions and services as easily as possible. Tooling should work intuitively. Period.
• You look at clouds We are migrating towards GCP, and you are a main driver in how we leverage Machine Learning tooling and apply that to our big dataset.

• You are a Data nOOb If you never worked with Data at scale, this might not be your ideal position.
• You are a single speed bike Nothing wrong with those (great for city commutes!), but to make the most of this role requires extra gears. Bol changes fast, and you need to thrive in a fluent environment.
• You do not care about sharing You lock knowledge and insights away in a safe. And throw away the key




Language requirements:

Specific requirements:

Educational level:

Level of experience (years):

Senior (5+ years of experience)

How to apply:

Please mention NLP People as a source when applying


Tagged as: , , , , ,