External reviews
189 reviews
from
External reviews are not included in the AWS star rating for the product.
CAST AI Makes Kubernetes Cost Optimization Truly Automated and Reliable
What do you like best about the product?
what i like best about cast ai is how it turns kubernetes cost optimization from a manual, reactive task into an automated, intelligent system.
instead of relying on static node groups or basic cluster autoscaler logic, cast ai continuously analyzes pod resource requests, actual utilization, bin-packing efficiency, and real-time spot and on-demand pricing. based on that data, it dynamically selects the most cost-effective instance types without manual intervention.
a few things that stand out:
workload-aware autoscaling that optimizes instance type selection, not just node count
strong spot optimization with stability controls for production workloads
clear visibility into over-provisioning and inefficiencies
reduced operational overhead compared to manually tuning node groups
in my experience using it in a production kubernetes environment, the biggest value is continuous optimization without compromising reliability. it feels less like a monitoring dashboard and more like an active control layer for infrastructure cost efficiency.
instead of relying on static node groups or basic cluster autoscaler logic, cast ai continuously analyzes pod resource requests, actual utilization, bin-packing efficiency, and real-time spot and on-demand pricing. based on that data, it dynamically selects the most cost-effective instance types without manual intervention.
a few things that stand out:
workload-aware autoscaling that optimizes instance type selection, not just node count
strong spot optimization with stability controls for production workloads
clear visibility into over-provisioning and inefficiencies
reduced operational overhead compared to manually tuning node groups
in my experience using it in a production kubernetes environment, the biggest value is continuous optimization without compromising reliability. it feels less like a monitoring dashboard and more like an active control layer for infrastructure cost efficiency.
What do you dislike about the product?
while cast ai delivers strong automation and cost optimization, there is a learning curve in fully understanding and trusting its automated decision-making, especially for teams used to managing infrastructure at a very granular level. in some cases, having deeper visibility into the exact reasoning behind instance selection or node replacement decisions would add even more confidence during audits or incident reviews. however, these are more about enhancing transparency and familiarity rather than limitations in capability, and with proper kubernetes configuration and governance, the platform performs reliably and efficiently.
What problems is the product solving and how is that benefiting you?
cast ai is solving the core problem of kubernetes cost inefficiency and over-provisioned infrastructure by continuously optimizing compute selection, bin-packing, and spot utilization in real time. instead of relying on static node groups or periodic manual reviews, it automatically matches workloads to the most cost-effective instance types based on actual usage and market pricing. this has reduced waste from idle resources, improved cluster efficiency, and minimized the engineering time spent on manual tuning and capacity planning. the biggest benefit for me has been shifting from reactive cost control to continuous, automated optimization while maintaining production stability.
Intuitive, quickly integrated, and noticeably saves time and costs
What do you like best about the product?
Very intuitive operation.\nEasy integration into existing systems.\nQuick implementation without much effort.\nGood overview in the dashboard.\nSaves time and costs.
What do you dislike about the product?
Partial lack of transparency in optimizations. Extended functions could be explained more clearly. Reporting could be more detailed.
What problems is the product solving and how is that benefiting you?
Automatically reduces cloud costs. Optimizes resources efficiently. Saves manual work in the DevOps team.
Automated Rightsizing That Cuts Cloud Spend Without Sacrificing Performance
What do you like best about the product?
The automated workload optimization and rightsizing capabilities are the most effective features. The platform’s ability to analyze resource requirements in real-time and automatically adjust instances ensures that the infrastructure remains efficient without requiring constant manual intervention from the engineering team. This lead to a measurable reduction in cloud spend while maintaining cluster performance.
What do you dislike about the product?
While the initial returns were simple to realize on lower environments, the transition to production required more effort. We found that iterative tuning was necessary to ensure there was no application degradation or performance impact. A more seamless "plug-and-play" experience for high-traffic production workloads would be a valuable improvement.
What problems is the product solving and how is that benefiting you?
Cast AI addresses the challenge of managing complex cloud infrastructure costs and resource allocation. By automating the rightsizing of clusters and optimizing instance selection, it removes the manual burden of infrastructure tuning from our engineering teams.
Clear Infrastructure Cost Insights with Cast Ai
What do you like best about the product?
Cast Ai provides a clear overview of my infrastructure costs. It helps me understand where the spend is going and makes it easier to identify the lowest prices.
What do you dislike about the product?
It has recently become more expensive. Maybe that’s due to scaling up or adding extra features we don’t need.
What problems is the product solving and how is that benefiting you?
It significantly lowers the cost of the nodes we need to run our jobs. We use it primarily for cost efficiency.
Beautiful Scaling Mapping and Read-Write View for Zero-Downtime Strategies with Less Cost
What do you like best about the product?
The best thing and the best feature of Cast AI is beautiful mapping of
1. nodes scaling
2. horizontal scaling
3. provides read only view so LLMs can learn and optimise the strategy instead of just implementing directly to our environment and learns itself 0 downtime strategies.
4. Their customer support is too good, For P0 issues Chandani from castAI is available on prompt basis
5. We can implement castAI just by providing the enough IAM permissions, and easily installing castAI into our EKS environment within 30 mins.
6. Our infra cost is reduced to 30% by using castAI for just like 40-50 days.
1. nodes scaling
2. horizontal scaling
3. provides read only view so LLMs can learn and optimise the strategy instead of just implementing directly to our environment and learns itself 0 downtime strategies.
4. Their customer support is too good, For P0 issues Chandani from castAI is available on prompt basis
5. We can implement castAI just by providing the enough IAM permissions, and easily installing castAI into our EKS environment within 30 mins.
6. Our infra cost is reduced to 30% by using castAI for just like 40-50 days.
What do you dislike about the product?
There is nothing to dislike, but there can be one improvement
We can have correct mapping if we are using nginx-ingress, as we have to map target groups of nginx ingress in castAI console.
We can have correct mapping if we are using nginx-ingress, as we have to map target groups of nginx ingress in castAI console.
What problems is the product solving and how is that benefiting you?
We donot have to bump to EKS console to view things, as castAI gives best user interface with extra capabilties and eliminated the need of having karpenter. It also eliminated the need of mapping nodeSelectors, affinity, taints, tolerations as we can manage them on castAI by just a go.
Easy-to-Use Tool with Useful Workload and Node Balancing
What do you like best about the product?
Workload balancing and node balancing are useful overall, and the tool is easy to use.
What do you dislike about the product?
It’s expensive. For now, it pays for itself, but it’s definitely not really cheap.
What problems is the product solving and how is that benefiting you?
uese AI to balance workfload
Node Type Flexibility and Rebalancing Make Maintenance Easy. And the customer support is Bravo
What do you like best about the product?
Being able to choose specific types of nodes has helped us a lot. I also love the rebalancing feature, which makes upgrades and ongoing maintenance much easier.
What do you dislike about the product?
Well the logs/audit-logs are a bit less informative to debug somethings.
What problems is the product solving and how is that benefiting you?
The biggest problem its solving is scaling right amount/size of nodes and with features like graceful draining.
Also I like the feature that we can have multiple node templates/configuration and that can be used using labels and selectors
Also I like the feature that we can have multiple node templates/configuration and that can be used using labels and selectors
Hands-Free Automation Cuts Costs Dramatically
What do you like best about the product?
I like the hands-free automation that actually executes changes rather than just giving advice. It eliminates manual toil by automating cost-saving infrastructure changes instantly. Before using CAST AI, our Kubernetes spend was a black box, and after connecting it, we discovered we were over-provisioned by 40%. The automated rebalancing cut our monthly bill by half within the first week.
What do you dislike about the product?
I feel it's costly. The setup complexity is an issue along with limited documentation and pricing. I wish there was better SMB pricing, clearer onboarding, and more granular policy documentation.
What problems is the product solving and how is that benefiting you?
CAST AI transformed our opaque K8s spending, revealing a 40% over-provision. Its automation executed cost-saving changes, halving our bill in the first week and eliminating manual toil.
Single Pane of Glass for All Our K8s Clusters
What do you like best about the product?
It gives me visibility into all the K8s clusters we manage and operate from a single entry point, so there’s no more clicking around between different dashboards.
What do you dislike about the product?
The current pricing model isn’t allowing us to roll out cast across all clusters, including our nonproduction environments.
What problems is the product solving and how is that benefiting you?
It provides cost visibility for both product teams and the infrastructure team into usage of a shared cluster. The technical utilization data also helps show where we can optimize.
A tool that changes the game
What do you like best about the product?
I appreciate the range of diverse resources that the tool offers and, in addition to the constant updates focused on improving the tool, it is possible to notice the appreciation of those who work on this project and their commitment to excellence. The ability to identify how much each team's application running on the cluster costs is sensational.
What do you dislike about the product?
Sometimes, its scaling is not that interesting. I don't know if this happens because of the calculations of Requests and Limits, but it often tries to spin up some nodes with absurd amounts of memory, which would make me spend much more than save. Therefore, it is necessary to impose several limits and not let it run loose; otherwise, the cost may end up increasing instead of decreasing.
What problems is the product solving and how is that benefiting you?
Currently, I am using CAST AI to control the automatic scaling of my cluster, as well as the vertical scaling of my workloads, focusing on cost reduction and operation optimization. This is possible thanks to the artificial intelligence of CAST AI in allocating nodes with the best machine type for optimized operation, performance, and cost. Additionally, I also use CAST AI to turn on and off my non-productive clusters, focusing on cost savings.
showing 31 - 40