Case Study
From Manual Bottlenecks to Scalable Precision in Health Tech

The challenge
A health technology startup needed to turn its research workflows into a first product. Large datasets were being processed on local machines, sometimes for hours or days, and people had to check files manually. That made each run demanding for the team and made it harder to handle more data.
What we did
Das Meta built infrastructure on AWS, providing cloud computing resources for the product instead of relying only on local machines. We built data workflows using Prefect, a tool for coordinating steps in a workflow, and Spark, a tool designed to work through large datasets across multiple computing resources.
We used Kubernetes, software that manages applications running in containers, to help run the workload. Containers are packages containing an application and what it needs to run consistently; EKS is AWS's managed Kubernetes service. Terraform describes infrastructure in files so that resources can be set up consistently. GitLab CI provided deployment pipelines: repeatable steps for moving software changes into an environment. Prometheus and Grafana collected and displayed operational information for the team.
Where that left the team
The initial six-month project was followed by a retainer. The case describes the infrastructure and data workflows built for the product; it does not establish a verified processing-time reduction. Explore cloud infrastructure management.
Frequently Asked Questions
What challenge did the health technology team face?
The team needed to improve data processing workflows that were affected by manual bottlenecks.
Which technologies are discussed?
The case study describes data workflows involving AWS, Kubernetes, Spark and Prefect.
Is the customer named?
The customer is presented anonymously in the case study.
