What I value most is how Cast AI turns complex Kubernetes optimization into a seamless, automated process. It consistently delivers significant cost savings (often up to 50%) by intelligently handling Spot instances and right-sizing workloads in real-time. The automated autoscaling is much more responsive than native cloud tools, and the 'set and forget' nature of the platform has effectively removed the burden of manual cluster tuning from our DevOps team. Review collected by and hosted on G2.com.

Initial Trust & Security Approval: Because Cast AI requires deep access to your cloud environment to perform automated actions (such as adding or deleting nodes), the initial security review—especially in larger organizations—can be lengthy and quite rigorous.

Learning Curve for Advanced Policies: While the “read-only” setup is fast to get up and running, getting comfortable with the more advanced automation policies—and correctly fine-tuning the “Auto-Remediation” settings—takes time and a strong understanding of Kubernetes to avoid unexpected behavior.Cast AI solves the persistent issues of cloud waste and operational complexity. Before, we spent too much time manually adjusting resources and overpaying for idle capacity. Now, the platform benefits us by automatically selecting the most cost-effective nodes without compromising performance. This hasn't just lowered our monthly bill; it has given our engineers the peace of mind and the time to focus on building features rather than managing infrastructure. Review collected by and hosted on G2.com.

I really like the vulnerability visualization feature that CAST AI provides; it identifies vulnerabilities and shows the vulnerable images, highlighting their criticality. I also really appreciate all the information it provides and the possibility of automations, which is very important for the organization and greatly eases the daily workload. Review collected by and hosted on G2.com.

One of the areas where CAST AI needs to evolve is in the matter of multiple organizations within the same company. Today, there is no centralized management, which makes daily operations quite difficult. If you have an environment that requires an organization, you need to create another one, which makes managing these tokens more difficult. Furthermore, there is no centralization in terms of a main organization and subsidiary organizations. This ends up multiplying the users, and they need to be managed manually, complicating the management of the tool. Review collected by and hosted on G2.com.

I really like how CAST AI works and appreciate that it benefits us by saving costs. Workload rightsizing is a unique feature that saves a lot of resources and costs. It's easy to configure and very flexible. Support is responsive. CAST AI is way ahead of tools like Karpenter, which are partly used under the hood but don't compare to CAST AI's offerings. The user interface is nice and customizable, which makes it convenient to use. It works well with Terraform and GitOps tools like ArgoCD. Using API for configurations alongside these tools adds to the seamless experience. Review collected by and hosted on G2.com.

One of our clients is on Azure and some of CAST AI's features seem to be made more for AWS and do not work well in the UI for Azure, but there are ways to make them work over API so it's a minor issue. Review collected by and hosted on G2.com.

CAST AI is most helpful because it automates Kubernetes optimization at scale, and the upside is measurable cost savings, improved resource efficiency, and reduced operational overhead. Review collected by and hosted on G2.com.

the downsides aren’t about capability — CAST AI does what it’s designed to do well — but around transparency, scope expectations, and fit. It works best when clusters are large, workloads are variable, and teams are comfortable embracing automation. Review collected by and hosted on G2.com.

Users of CAST AI appreciate the shift away from manual “firefighting” toward higher-level automation, especially features like Scalers and Performance. At the same time, some users place less value on the User Interface and Dashboards. Review collected by and hosted on G2.com.

Lack of professionalism at the support level, lack of knowledge about how the components work and what they are responsible for, lack of logs, disorganized version deployment. No partner updates regarding changes to features and versions themselves. Changing feature definitions in new versions without updating the client. Full of bugs in Terraform's code. Review collected by and hosted on G2.com.

I like how CAST AI calculates the cost on a daily basis and provides insightful financial granularity. This helps me catch spending spikes in real time and manage my services effectively to prevent bill shock. The new console UI is friendly and easy to use, which simplifies handling complex Kubernetes data. I love how it turns these complexities into a one-click executive dashboard for effortless FinOps management. Setting it up was very easy, even without much previous experience, making my transition to using CAST AI smooth and problem-free. Review collected by and hosted on G2.com.

Sometimes it's hard to believe in the product and the trustworthiness of the data it provides. More cost-accurate calculations are needed. Review collected by and hosted on G2.com.

I like being able to see everything on the dashboards and make decisions right there with CAST AI. These dashboards are a very good tool for resource identification and optimization, allowing me to quickly visualize where I need to focus. Review collected by and hosted on G2.com.

I feel a bit insecure about putting CAST AI in automatic mode because, although it is simple, it can affect production. There is no intuitive and quick way to ensure it, and I feel like I have to set many rules to prevent it from downscaling things it shouldn't for some reason. Also, the initial setup gave us quite a struggle because we use Terraform and had to deal with permission issues, among other things. Review collected by and hosted on G2.com.

With Cast AI, I can clearly identify the amount of CPUs and memory that my cluster is using, as well as the cost it is generating. Additionally, the tool allows for very efficient and secure resource optimization, providing more control and visibility over the environment's usage. Review collected by and hosted on G2.com.

No disadvantages. Since we started using Cast AI in our environments, it has helped us a lot in cost optimization. Review collected by and hosted on G2.com.

The reliability of the node scaling continues to be a highlight; we have zero anxiety about node recreation even during high-traffic events. The 'set it and forget it' nature of the infra management is still the strongest selling point. Stability has been rock solid over the last year. Review collected by and hosted on G2.com.

I still find the cost reporting to be the weak point compared to the GKE/GCP console. I need better visibility into specific node template costs and a clearer breakdown of infrastructure prices. Currently, reconciling Cast AI's reported figures with our actual GCP billing lacks the necessary granularity. We need more transparency on the total amounts for specific node groups to fully trust the cost data. Review collected by and hosted on G2.com.

I like CAST AI for its workload autoscaling, which helps with optimization while maintaining availability. It has helped me achieve about 70% cost optimization with better bin packing, rebalancing, and pod mutation to maintain a diverse workload. I also appreciate having a single place to manage multiple clusters and daily costing. The initial setup was very easy and well-guided by the CAST AI team. Review collected by and hosted on G2.com.

I think it could be better for handling StatefulSets in bin packing. It can handle better StatefulSets to make sure pods should handle state while eviction and reschedule in other nodes. Review collected by and hosted on G2.com.