Senior Data Engineer, AI​/ML Platform

Location:
Amsterdam (hybrid) | Full-time

Roamler is a Dutch technology company that helps clients grow their business, improve end-customer experience, and reduce and flexibilise the cost of field operations. It combines actionable insights with the execution capabilities required to turn those insights into measurable results. Roamler operates through two business units:

• Roamler Retail provides in-store and out-of-home insights and execution for FMCG brands and retailers across Europe, enabling clients to increase sales volumes while optimising operational efficiency.

• Roamler Tech delivers installation, maintenance, and repair services for large B2C service providers, improving customer experience while increasing the efficiency and flexibility of field execution.

With approximately 130 employees and a large network of flexible workers active across multiple European markets, Roamler supports clients including P&G, Friesland Campina, Unilever, and Eneco.

Founded in Amsterdam in 2011, Roamler is backed by Endeit Capital and Smartfin, and operates across the Netherlands, Belgium, Germany, the United Kingdom, France, and Spain.

We are seeking a Senior Data Engineer to join our AI/ML platform team and take ownership of the pipelines and systems behind our outlet‑data product suite, the data platform behind our OOH Location Database, Salesmapp, and related products used by food & beverage brands across Europe.

You’ll work closely with our data scientists, who own the classical ML/NLP enrichment pipeline (matching, classification, embeddings), as well as modern LLM and AI pipelines. Solid working knowledge of this domain is a genuine plus, since your systems feed directly into it.

• Own and evolve our data ingestion pipelines and our web scraping engine as production systems: architecture, reliability, scalability, and performance, not just individual scripts.

• Design and build AI agent based extraction workflows (for example browser automation agents that navigate outlet websites, menus, and listing platforms) where they outperform traditional scraping and parsing approaches.

• Build evaluation frameworks and quality checks so AI‑driven extraction and enrichment can be trusted at the same bar as our existing rule‑based systems.

• Utilize AWS & Azure services effectively across both the traditional pipeline (Airflow, Spark, Glue) and the AI‑driven components (model hosting, agent orchestration).

• Enhance our automated testing, data validation, and monitoring capabilities, including for AI components, which introduce new failure modes and cost profiles compared to deterministic scraping and rule‑based logic.

• Collaborate with business and data stakeholders, including our data science team, to align requirements and expectations across the full pipeline.

• Drive engineering projects end to end and ensure successful outcomes.

• Drive continuous improvement in our engineering setup.

• Foster a collaborative and enjoyable team environment through code reviews, knowledge sharing, and daily interactions.

• Bachelor's and/or Master's degree in computer science or a related field.

• 5+ years of experience as a Data Engineer with strong software engineering skills, including ownership of production data pipelines and/or web scraping and crawling systems.

• Design, build, and operate scalable data pipelines on AWS that bring in data from external web sources and turn it into clean, queryable datasets with Playwright and/or Beautiful Soup.

• Experience with Airflow, Python, SQL, and Spark.

• Hands‑on, practical experience integrating LLMs into production systems: prompt design as part of system design, evaluation, and cost/latency tradeoffs, not just experimenting in a notebook.

• Experience with AI agent frameworks (for example browser automation agents) or a strong interest in and aptitude for learning them.

• Manage infrastructure as code with Terraform on AWS (ECS, EMR, Glue, S3, SQS/SNS, IAM)

• Develop and maintain Spark jobs on EMR for batch ETL and enrichment at scale.

• Some exposure to classical ML/NLP concepts (classification, embeddings, matching) is a plus, since our data science team's pipeline…