UnderTheHood at Cowrywise

​UnderTheHood is a series we run at Cowrywise where our engineers break down how something on our platform was actually built. We talk through the decisions we made, the trade-offs we accepted, and the parts that did not go smoothly. Then we open the room and invite attendees to question the architecture and challenge our approach.

​We’re bringing it back this year, starting with a session led by Chima Ataman from our Infrastructure team.

​The focus is how we handle orchestration on Docker Swarm at Cowrywise.

​As our services grew, cracks began to show. Node setup was slow. Auto Scaling Group refreshes occasionally took services down. ECR credentials expired mid-deploy. What worked at one stage did not hold up at the next.

​Chima will walk through how we rebuilt the system by combining custom AMIs, Auto Scaling Groups, CodeDeploy, private networking, and GitHub Actions into a setup that now allows developers to deploy without worrying about infrastructure.

​If you work in infrastructure, DevOps, or backend engineering and you’re curious how a fintech team handles orchestration at scale, this session is for you.