A startup founded by former Uber engineers launched an autonomous infrastructure optimization platform it claims can refine Kubernetes workloads to help slash ballooning compute and inference costs.

Dubbed DevZero, the startup of the same name said its platform can optimize Kubernetes at the cluster, node, and workload level “instantly.” Without the need for restarts, DevZero claims the offering can be used to reduce compute bills by up to 60% while also improving workload efficiency and reliability.

“Infrastructure teams are hesitant to let anything manage compute autonomously because that usually comes with tradeoffs. Cutting the cloud bill isn’t worth it if the result is downtime,” DevZero CEO Debosmit Ray explained. “We built DevZero to eliminate the tradeoffs between performance, reliability, and cost. This is a platform for engineers who want autonomous optimization they can trust at 3 am.”

Ray co-founded DevZero alongside Uber alum Rob Fletcher, with the pair having worked as engineers building automation tools for the ride-share giant. While their startup began as a cloud development platform to improve coder productivity, it pivoted to Kubernetes and autonomous infrastructure optimization to help reduce compute over-provisioning and overspending.

The pair contend that manual rightsizing of workloads is costly and incredibly labor-intensive, while rival autonomous systems are “rarely implemented” with teams believed to be “justifiably nervous” to hand control to autonomous systems over fears such offerings could under-provision compute and thereby ruin vital applications.

DevZero is positioned with trust in mind, packing a profiler that monitors clusters, nodes, and individual workloads that feed into building statistical models of resource demand so engineering teams can better understand their workloads. The platform itself uses those models for context-aware scheduling and autoscaling layering, placing workloads efficiently to ensure cost-effective capacity.

The startup said its flagship platform can rightsize workloads in real time, adjusting central processing unit (CPU), memory, and graphic processing unit (GPU) provisioning, while also ensuring live migration in the case of demand spikes or availability zone outages.

Customers already on board include AI development platform Outerbounds (recently acquired by Anaconda), health care procurement platform Dentira, and data pipeline platform DataBahn.

“During a recent availability zone outage, DevZero transparently migrated our workloads live without requiring a single restart or operational intervention from our team,” Mihir Nair, DataBahn’s head of architecture, revealed. “That level of resiliency gave us the confidence to push infrastructure optimization much more aggressively. We’re now partnering closely to extend that operational intelligence further with Kubernetes observability for outbound network traffic optimization, reduced data transfer costs, and continuous security posture monitoring.”

DevZero has raised $26 million in funding, per Crunchbase figures, securing backing from Anthos Capital, Foundation Capital, and Madrona Ventures.