Orchestrating data workflows using a fully serverless architecture

Fundbox is a growing fintech company that provides an automatic underwriting platform based on data and AI. While scheduling a limited number of data workflows is a generally manageable task, scaling to hundreds of data workflows with dependencies and diverse job types requires substantial customized engineering, complexity, and overall expensive resources. Serverless-based architectures offer an alternative to traditional resource management.

Tomer Levi explains how the data engineering team at Fundbox uses AWS Step Functions, Docker containers, and Spark to build a live, serverless data orchestration platform, focusing on the company’s decision to build a friendly, yet powerful and scalable solution. Tomer details AWS Step Functions state machines, their limitations, and how to overcome them by building custom job-scheduling and dependency features. He illustrates how resource bottlenecks were overcome using Docker containers and AWS Fargate. Fundbox’s architecture is scalable and already serves dozens of engineers, BI developers, and data scientists in the company.