How Kapstan helped Hakimo boost reliability and cut cloud cost by 40%

Shyam Kumar
Shyam Kumar
In this article

What you'll learn

  • How Kapstan cut Hakimo’s cloud spend by 40% in a month
  • Transitioning to multi-cluster Kubernetes set up to improve stability across environments

Hakimo is a leading AI-powered physical security platform focused on remote guarding and computer vision. Hakimo has been scaling rapidly; their primary focus was been customer acquisition, AI product development, and partner integrations. As a result, their infrastructure and cost management were de-prioritized. However, as their operations scaled rapidly, those deprioritized items began to add up to substantial overhead costs.

Challenges Faced

Hakimo has a world-class engineering team, but they faced a number of operational challenges and resource constraints that led them to seek external expertise. The biggest reason? Cloud costs were spiraling out of control. Despite the drastic rise in cloud spend, the Hakimo engineering team needed to remain laser focused on customer commitments and product roadmap priorities. As a result, important DevOps related work including multi-cluster Kubernetes management and cost optimization constantly took a back seat.

Kapstan Solution

Hakimo partnered with Kapstan to address these issues holistically. The collaboration began with a Proof of Concept (POC) focused on optimizing AWS infrastructure to reduce operational costs and ensure scalability.

  1. 40% cloud bill reduction in 40 days

Kapstan recommended a number of items to optimize resource allocation and architecture requirements. We wanted to ensure Hakimo had the best set up first so they could see immediate cost savings, before migrating them into Kapstan. Below is a list of items we tackled:

  • VPC flow log reduction
  • Right sizing various infrastructure pieces such as RDS and OpenSearch
  • Optimizing Queuing infrastructure to minimize empty message retrievals
“Kapstan’s expertise has been instrumental in helping us reduce AWS costs and scale our platform. Kapstan took the infrastructure management burden off our shoulders, enabling our team to focus on core AI development”  – Amit, Head of Engineering at Hakimo
  1. Infrastructure Scaling

Hakimo utilized Kapstan’s expertise to transition from a single Kubernetes cluster to a more scalable and reliable multi-cluster setup, which greatly enhanced stability and maintainability. By leveraging Kapstan’s environment management feature, Hakimo achieved a clear separation of resources for operational stages, such as development and production, effectively avoiding conflicts. Additionally, Kapstan’s implementation of automation gates ensured consistency and reliability across deployments, further boosting operational efficiency.

  1. CI/CD Implementation
    Kapstan worked closely with Hakimo to implement a seamless CI/CD pipeline, improving developer productivity and code deployment efficiency. By automating the integration and delivery processes, Hakimo’s development team can focus more on innovation rather than managing deployments.

Future Plans

Hakimo continues to work with Kapstan to further utilize Kapstan’s out of the box observability features. The next phase of the partnership will focus on refining these areas to ensure complete visibility into system performance and efficiency. Hakimo strongly believes in “Continuous compliance” and looks forward to working with Kapstan for SOC2 journey.

Conclusion

Kapstan has been instrumental in Hakimo’s journey, simplifying DevOps management, prioritizing critical features, and embedding best practices. This partnership allowed Hakimo to focus on their core mission — AI-powered security monitoring —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.  

Simplify your DevEx with a single platform

Schedule a demo