While working as Senior Software Engineer and Enterprise Architect for apilayer GmbH, I was invited to help with the Marketstack project. It was a short-term appearance, but the product and challenges were complex and unique enough to add it to other portfolio items.
Marketstack was designed to gather 170.000 worldwide market tickers in real-time and then serve this data via API to thousands of customers. So it is understandable that scaling was one of the challenges once the product got some traction. My task was to investigate data models and to fetch procedures with the marketstack development team.
Together, we quickly found bottlenecks and proposed conceptual improvements to the data model which were feasible for implementation. These changes resulted in significantly faster data queries. In addition, it immediately resulted in faster and more stable API and reduced strains on application and data servers.
In short, the critical impact for the product itself was stability, scalability, and significant performance increase based on small-scoped but targeted changes. In addition, this project was a fantastic learning experience for me. I could look under the hood of what was designed to be a high-intensity data-driven system and understand what was considered during the architectural software planning.
Marketstack is a product that delivers real-life stock data via REST API interface in JSON format. With more than 170.000 tickers, they are trusted by 30.000+ customers.
My role: Senior Software Engineer, Enterprise Architect
Impact: Performance optimization and scaling, focusing on data banking aspects.