“Without Kapstan, start ups have to go through a steep DevOps learning curve. Kapstan’s expertise takes away our day-to-day headaches, allowing our team to focus entirely on what our users care about.” — Abhishek Mishra, Founding Architect, Headlamp
About Headlamp
Headlamp Health is a data management solution bringing precision data to mental health. They’re doing this with a dual sided platform: an application that equips mental health clinicians with advanced tools to analyze comprehensive mental health data, and a second application that gives users access to their Universal Health Records, and collects longitudinal information. These two applications feed data to a data management platform that allows psychiatrists to create and administer more personalized and effective treatment.
Background
Although Manaswini (Head of Engineering) and her founding team were well-versed in Kubernetes, they initially considered serverless and PaaS options like Vercel to quickly bring their product to market. While they knew Kubernetes offered power and flexibility, its complexity would have distracted them from core product development. They also recognized that eventually, they would need Kubernetes as their product scaled. Knowing Kapstan could simplify Kubernetes management without sacrificing its benefits, they chose Kapstan, allowing them to focus on building their product while leveraging Kubernetes’ full potential.
Even as their team scaled, and onboarded new engineers, they didn’t have to worry about learning Kubernetes or worry about DevOps.
The Journey
As Headlamp’s engineering team grew, so did the number of applications they managed. At the start of their journey with Kapstan, they had four services in production. Over the next 12 months, they added 20 more services and expanded their engineering team from 4 to nearly 20 engineers. While the founding team had extensive infrastructure knowledge, many of the new engineers were less familiar with the complexities below the application layer.
Kapstan removed this challenge entirely. Spinning up new services became as simple as a few clicks on the UI. Engineers could quickly launch a service in the staging environment, test it thoroughly, and seamlessly promote it to production. This ease of use allowed the team to focus on delivering business value without needing to hire dedicated DevOps personnel or worry about the steep learning curve of Kubernetes.
“I wanted to focus our teams' efforts on the business. Behind the scenes, I didn’t want to worry about not being able to scale, managing different issues in all our k8s clusters, and fixing our build pipeline. In my previous company, we had several people who were solely responsible for doing this. But with Kapstan, those problems are automatically taken care of. -Manaswini, Head of Engineering
Key benefits of using Kapstan
A few key things that elevated the Developer Experience of Headlamp engineering team are the following:
- Simplified Kubernetes management: New engineers at Headlamp didn’t need to earn a Ph.D. in Kubernetes to manage production-scale infrastructure. Kubernetes monitoring, routine maintenance like cluster upgrades, cost optimization, and auto-scaling (Karpenter) were all built-in, and easily controllable via an easy-to-use UI. This empowered the team to operate independently without deep technical knowledge around kubernetes.
“Kapstan has simplified deploying, managing and scaling applications on Kubernetes, especially for users who might not have a deep background in Kubernetes itself. It abstracts some of the complexity of Kubernetes, includes tools for monitoring and troubleshooting that are designed to be easier to use, providing a more user-friendly interface for deploying and managing applications.
It often includes default configurations that simplify the deployment process, allowing you to get started quickly rather than dealing with Kubernetes intricacies.“ — Ankit, Software Engineer, Headlamp
- Built-in security and compliance: Security and compliance are paramount in the healthcare industry. Kapstan ensures that deployed infrastructure is always up to date with latest security patches and code is scanned to identify vulnerabilities. Secrets are securely stored in a vault and are rotated periodically.
- Prioritized feature development: Kapstan evolved alongside Headlamp, adding features like environment cloning and cron job support, allowing Headlamp to scale without the burden of managing complex DevOps.
“Kapstan’s team was responsive to our needs, making it easier to prioritize critical infrastructure improvements without compromising on business requirements.” — Manaswini, Head of Engineering, Headlamp
- Out-of-the-box observability: Kapstan plays nice with other observability tools such as DataDog, but the Headlamp engineering team opted to give Kapstan’s built-in monitoring+observability solution a shot. As soon as engineers deploy a new service via Kapstan, applications are automatically set up with proper alerting, monitoring, and cost observability. No more defining sidecars, modifying YAML files, instrumenting the application, configuring the side car for proper data collection, and ensuring that the sidecar communicates with the rest of your monitoring stack. The Headlamp team opted to leverage Kapstan’s built-in solution, instead of paying a premium for DataDog.
Kapstan 24/7 Support
DevSecOps isn’t a one-time task—it’s an ongoing effort that grows more complex as your product evolves and your team scales. As new services are added, the challenges of managing cost, reliability, scalability, and security also increase. Startups often face the need for constant oversight and expansion of platform engineering teams to handle these complexities. Kapstan steps in as an extension of your engineering team, providing continuous support to ensure everything runs smoothly.
A few examples of Kapstan’s timely and customized support for Headlamp include:
- Kubernetes Event-Driven Autoscaling (KEDA): Headlamp Health’s services heavily rely on Kafka queues. At the time, Kapstan only supported KEDA with SQS. However, within just 5 days, the Kapstan team extended KEDA support to Kafka, ensuring that Headlamp’s infrastructure remained scalable and efficient.
- Cron Job Support: Headlamp needed a reliable way to run cron jobs. Kapstan simplified the entire process, not only making it easy to configure, but also providing visibility into job history—whether the job ran successfully or failed. With Slack integration, Headlamp’s team receives real-time notifications, eliminating the need for manual monitoring and offering peace of mind.
Results
With Kapstan, Headlamp Health has zero DevOps toil. The automation that Kapstan provides Headlamp Health allows them to:
- Efficiently leverage resources: Kapstan allowed Headlamp to allocate more resources to product development. If it weren’t for an internal developer platform like Kapstan, Headlamp would have had to find, train, and onboard more platform engineering members early on in their journey, taking away crucial resources from building their product.
- Scale smartly: Headlamp could handle user growth seamlessly, with infrastructure scaling managed effortlessly through Kapstan’s built-in Kubernetes Event Driven Architecture.
- Manage cost: The platform’s cost monitoring and recommendation tools enabled Headlamp to save over $20k/year that would have otherwise been spent on DataDog and other tooling.
Conclusion
Kapstan has been instrumental in Headlamp’s journey, simplifying DevOps management, prioritizing critical features, and embedding best practices from day one. This partnership allowed Headlamp to focus on their core mission—enhancing mental health care—while Kapstan took care of their DevOps needs.
If you’d like to learn how Kapstan can help do the same for your team, reach out to talk to our founding team.
