Cast AI - EKS fully automated cost optimization and monitoring

What do you like best about the product?

Running eighteen containerized microservices on Azure Kubernetes Service (AKS) for our insurance claims pipeline used to mean paying for a lot of idle cloud capacity. While we offload bursty GPU batch jobs to Modal Labs, our core 24/7 services-including our FastAPI api-server, orchestrator-service, cache-manager, and separate workers for transcription, mapping, categorization-worker, pricing, depreciation-worker, and tax-all live on Kubernetes. What I like best is how easy it was to onboard our clusters using their Terraform provider in read-only mode first, inspect the savings recommendations, and then turn on the Workload Autoscaler and Spot instance automation. Through the Cast AI console, I was able to assign our stateless categorization-worker and depreciation-worker pods to run on cheap Azure Spot VMs, while keeping stateful services like orchestrator-service and our Kong Konnect gateway pods pinned to stable on-demand nodes. It handles pod rightsizing and node bin-packing automatically without my team having to rewrite Kubernetes YAML files every week.

What do you dislike about the product?

Turning on automated node rebalancing for the first time caused a few unexpected disruptions in our staging environment. Before we tuned our settings, Cast AI aggressively drained underutilized nodes to bin-pack pods and accidentally evicted our cache-manager and active orchestrator-service containers right in the middle of processing long multi-sheet claim exports. I had to go back into our Helm charts and manually add safe-to-evict: false annotations along with strict PodDisruptionBudgets so long-running jobs wouldn't get cut off. Also, when the Workload Autoscaler adjusts CPU and memory requests vertically, it can trigger rolling pod restarts unless you lock those updates to an overnight maintenance window. Finally, the dollar savings shown in the dashboard are calculated against public cloud list prices, so reconciling those numbers with our discounted Azure enterprise invoice takes some extra spreadsheet work.

What problems is the product solving and how is that benefiting you?

Sizing our Kubernetes pods used to be complete guesswork. Overnight, our pipeline containers often sat around 10% CPU usage while holding onto huge memory reservations, wasting money on oversized Azure VMs. Then, whenever a severe storm triggered a wave of catastrophe (CAT Code) inventory claims, our pricing and depreciation workers would suddenly hit memory limits and crash with OOMKilled errors. CAST AI solved both problems at once. It shrinks our cluster footprint automatically during quiet hours and spins up right-sized Spot nodes in seconds when adjusters upload heavy contents spreadsheets. We cut our monthly Azure Kubernetes compute bill by roughly 45% and stopped out-of-memory crashes across our worker queues.

What do you like best about the product?

The best part of Cast AI is its automated scaling and resource optimization features, which make it easier to manage Kubernetes infrastructure without spending too much time manually adjusting resources. The dashboard also provides a clear overview of resource utilization, which is especially helpful when working on cloud-based projects.

What do you dislike about the product?

For a beginner, a few advanced features may take some time to understand. Also, the documentation is quite complicated, which makes it harder to get started.

What problems is the product solving and how is that benefiting you?

It helps with my project by addressing unnecessary cloud spending caused by underutilized resources. It also provides better visibility into resource usage and helps optimize scaling, which reduces manual work.

What do you like best about the product?

I use CAST AI to automatically scale EKS and GKE resources and save money. It saves me the effort of adjusting Kubernetes resources dynamically and scaling the underlying infrastructure to meet pod needs. I like that it collects and aggregates useful metrics to adjust configurations; those metrics help me decide what to do next and plan ahead. CAST AI also automates capacity adjustments, and I find the UI onboarding easy and straightforward.

What do you dislike about the product?

I find the workload autoscaling powerful, but configuring it to meet the needs of different workloads can be hard, especially for job-like Pods. The UI onboarding is easy, but using Terraform for large-scale deployments is a little troublesome and requires more human effort. I’d like a generator or importer to help with Terraform onboarding.

What problems is the product solving and how is that benefiting you?

I use CAST AI to automatically scale EKS and GKE resources to save money. It saves me the human effort of dynamically adjusting Kubernetes resources and capacity to meet pod needs. Its aggregated metrics help me decide my next move, and many steps that used to be manual are automated.

What do you like best about the product?

What stands out most about CAST AI is how genuinely hands-off and dependable its automated Kubernetes cost optimization feels. Rather than only surfacing recommendations in a dashboard that engineers still have to implement manually, the autoscaler actively rightsizes workloads, bin-packs pods efficiently, and provisions just-in-time compute while avoiding cluster instability. Its handling of Spot instances is another big highlight. The automated fallbacks and graceful drain mechanisms let us confidently use Spot pricing for production-adjacent and batch workloads without the constant worry of downtime or unexpected node evictions. On top of that, the read-only savings report you get immediately after connecting a cluster provides clear, transparent visibility into idle compute waste before turning on full automation.

What do you dislike about the product?

The automated autoscaler can sometimes be overly aggressive with resource limits, which leads to pod churn or minor throttling during rapid traffic spikes. To mitigate this, I end up spending extra time fine-tuning disruption budgets and setting appropriate safety headroom. There’s also some friction between making manual adjustments in their dashboard and keeping everything consistent when managing state through Terraform. During onboarding, granting deep, cluster-level RBAC permissions can take a while to get approved by security teams. Lastly, the cost model doesn’t feel very economical for smaller workloads that don’t already have significant cloud compute waste.

What problems is the product solving and how is that benefiting you?

Before adopting CAST AI, our Kubernetes clusters were running with a lot of idle, over-provisioned compute, and our team spent countless hours guessing resource limits and manually tweaking YAML files. CAST AI automates this end-to-end by continuously right-sizing pods, bin-packing nodes more efficiently, and safely using Spot instances with automatic on-demand failovers whenever capacity drops. The biggest benefit for us has been a clear reduction in our monthly cloud compute bill, without introducing downtime or performance degradation. Just as importantly, it takes the day-to-day burden of capacity planning off our engineers’ plates, so we can focus on shipping software instead of constantly babysitting cluster resources.

What do you like best about the product?

The cluster looks healthy, but I’m still digging into what’s actually behind that green status. I find myself jumping between workloads, tracing resource behavior, and gradually the performance picture starts to make sense. I can take a quick look at the user interface, make a small adjustment, then go right back to the workload, and that rhythm genuinely suits the way I troubleshoot infrastructure.

What do you dislike about the product?

The onboarding feels too basic for an experienced engineer, with explanations that cover things I already know. On top of that, one integration behaves differently than I expected, which adds an unnecessary pause and interrupts the workflow.

What problems is the product solving and how is that benefiting you?

We were constantly chasing wasted capacity across our Kubernetes clusters, and our workloads needed manual tuning whenever traffic shifted. Cast AI started giving us AI-driven recommendations for resource allocation, and the ROI became clear once we applied those changes and began cutting unnecessary cloud spend.

What do you like best about the product?

I like CAST AI for its innovative approach to multi-cloud GPU capabilities, which allow me to choose GPUs across multiple cloud providers and even on-prem. This feature has been particularly beneficial for us. I also appreciate the resource optimizations, with memory right-sizing helping us identify workloads that need additional capacity. The automated workload optimization and extra insights on GPU optimization are also valuable. Additionally, the setup was very easy, and the customer support from CAST AI has been great.

What do you dislike about the product?

I think one area where CAST AI could improve is by increasing the transparency of its optimization decisions. It would be really helpful if the platform could offer richer insights into CPU and memory performance, especially around memory kills. This would allow us to proactively right-size the nodes. Having dashboards, alerts, and data-driven usage insights would also be beneficial. For example, understanding how minimum and maximum node settings impact applications that can have irregular operations throughout the month would be great.

What problems is the product solving and how is that benefiting you?

I use CAST AI for automated workload optimization and resource right-sizing. It helps identify workloads needing more capacity and offers insights on GPU optimization across multi-cloud providers.

What do you like best about the product?

One of the strengths of CAST AI is its cost optimization feature. I really like how it reduces costs using bin-packing technology and monitors available resources worldwide to operate at the most optimal cost.

What do you dislike about the product?

Due to different infrastructure configurations in each company, security settings were not easy. In particular, more integration capability is needed to resolve conflicts with security solutions. The initial setup was fine in the development network, but it took time to resolve firewall or security solution conflicts in the production network.

What problems is the product solving and how is that benefiting you?

The CAST AI in use maintains system stability while optimizing costs. It allows monitoring of available resources worldwide at the most efficient cost across various infrastructure configurations.

What do you like best about the product?

Its automated Kubernetes cost optimization which helped reduce cloud spend, improves resource efficiency, and removes much of the operational overhead of managing cluster scaling and infrastructure.

What do you dislike about the product?

The biggest downside is the learning curve, along with the need to build trust in the automation. Teams may need some time to understand how the optimization decisions are made and how to configure the platform in the best way for their environment.

What problems is the product solving and how is that benefiting you?

Cast AI addresses the challenge of Kubernetes cost optimization and infrastructure management. It helps identify overprovisioned resources, automate scaling, optimize node usage, and take advantage of Spot Instances to reduce cloud spend. As a result, we spend less time manually tuning clusters, get stronger resource utilization, and have clearer visibility into our cloud costs while still maintaining application performance.

What do you like best about the product?

Automatically cuts our Kubernetes costs without hurting speed.

What do you dislike about the product?

The UI is getting crowded. With security scanning, rightsizing, and spend forecasting added, finding a simple autoscaler setting can take three menus, and the docs seem to lag behind new features. I've had to contact support a few times, though they do respond. Node Templates also have a steep learning curve. You need to understand tolerations and node affinities to keep sensitive workloads like Redis and API gateways off spot nodes. Early on, a rebalance during a traffic spike briefly slowed our Retool dashboard. The savings-based pricing is fair to me as an engineer, but it's harder to explain to finance than a fixed license.

What problems is the product solving and how is that benefiting you?

Our biggest problem was a Kubernetes bill we couldn't explain. We run self-hosted inference, vector databases, and background jobs behind n8n and Retool, and I over-provisioned out of fear of latency spikes and cold starts. High-memory nodes and GPUs ran around the clock while sitting idle most of the time. Tuning requests and limits by hand was painful, and the native cluster autoscaler was too slow.
CAST AI fixed this in three ways. First, it showed us the waste. Our GPU utilization was in the single digits, and the projected savings were 40-50%. Second, it made spot instances usable. Batch document processing, transcription, and embedding jobs now run on cheap spot nodes, and it drains nodes gracefully before interruptions, so we don't get failed jobs or lost requests. Third, its bin packing keeps consolidating workloads onto the cheapest nodes, so we're not paying for idle capacity.
The result is lower compute costs, and I get my time back. I no longer babysit Grafana dashboards or explain why our GCP bill doubled after deploying a new model. Our real-time inference endpoints stay responsive too, as long as pod disruption budgets are set up properly.

What do you like best about the product?

Before using it, our engineering team spent hours manually adjusting node groups and resizing instances. Cast AI completely automated this workflow through its real-time autoscaling and advanced bin-packing algorithms. The UI/UX is incredibly clean and made the onboarding process seamless—we were able to run a read-only analysis first to see exact potential savings.

What do you dislike about the product?

The main area for improvement is the initial trust barrier and configuration complexity when setting up fully automated policies for mission-critical production workloads. Letting an automated platform dynamically manage node scaling and spot instances can be daunting for large-scale enterprise environments. While the read-only onboarding dashboard is excellent, configuring the fine-grained mutation policies to ensure specific pods never experience disruption requires a steep learning curve. Additionally, while their core Kubernetes cost optimization is top-tier, the visibility and optimization tools for non-containerized cloud infrastructure or managed services (like databases and storage) could be more robust. Expanding their AI intelligence to offer deeper suggestions for idle non-K8s resources would make it a truly complete cloud financial management tool.

What problems is the product solving and how is that benefiting you?

Cast AI is solving the massive problem of cloud waste and the operational overhead associated with managing Kubernetes infrastructure. Before adopting the platform, our engineering team spent significant time manually monitoring cluster utilization, resizing node groups, and trying to safely utilize spot instances. Cast AI has turned this into a completely autonomous process. The primary benefit has been a direct 50%+ reduction in our monthly compute spend. Beyond the immediate financial ROI, it has freed up our DevOps engineers from reactive infrastructure tweaking. They can now focus on product feature development rather than cloud cost engineering. It has bridged the gap between our finance team's demand for budget control and our engineering team's need for high performance.