Arrikto

Name: Arrikto
Brand: Arrikto

Arrikto provides infrastructure software for running production machine learning workflows on Kubernetes, including an enterprise Kubeflow distribution, a managed Kubeflow service, and a Kubernetes-native data management layer for reproducibility and portability.

Est 2014|Valuation $25M|Raised $14.8M|32 Ref Customers

LinkedIn 5K|Website|FTEs 6

MarketsWorkflow Orchestration

01Competitive Landscape

Narrative analysis

Areas of strength - Kubernetes-native and hybrid posture: EKF is designed for cloud and on-prem deployment, and Rok targets hybrid data distribution patterns. - Developer workflow integration: Kale provides a notebook-centric path to pipelines and integrates with Kubeflow components such as Katib and serving. - Emphasis on open-source tooling: Arrikto positions EKF around Kubeflow and open-source components to reduce ad hoc integrations. Where alternatives often have advantages - Managed cloud platforms (SageMaker, Vertex, Databricks) may be preferable when buyers prioritize reduced cluster operations and accept cloud provider lock-in. - Enterprise Kubernetes platforms (for example, OpenShift AI) may be more compelling when an organization standardizes on a single supported enterprise Kubernetes stack and prefers a single commercial vendor for support and subscription.

	Arrikto (EKF / KFaaS)	AWS SageMaker	Google Vertex AI
Starting Price	KFaaS: public usage-based rates for running and stopped deployments were listed; EKF: not public.	Usage-based; published by AWS but complex across features.	Usage-based; published by Google Cloud; varies by service.
Recommended use case	Teams committed to Kubernetes/Kubeflow; hybrid/on-prem deployments; organizations that want an open-source-based platform packaged for enterprise use.	AWS-native teams seeking a managed ML experience with integrations across AWS services.	GCP-native teams seeking managed MLOps components integrated with Google Cloud infrastructure.
Key strength	Kubeflow-centric portability; GitOps-oriented reproducibility tooling; Rok data layer and Kale for notebook-to-pipeline workflows.	Broad managed feature set and deep AWS integrations.	Integrated GCP ML services and managed MLOps components tied to Google Cloud.
Key weakness	Limited public third-party review evidence; continued operational responsibility around Kubernetes; EKF pricing not transparent.	Potential vendor lock-in and complex cost management.	Potential vendor lock-in and service sprawl; optimal experience typically within GCP.

01Competitive Landscape

Narrative analysis

Areas of strength

Kubernetes-native and hybrid posture: EKF is designed for cloud and on-prem deployment, and Rok targets hybrid data distribution patterns.
Developer workflow integration: Kale provides a notebook-centric path to pipelines and integrates with Kubeflow components such as Katib and serving.
Emphasis on open-source tooling: Arrikto positions EKF around Kubeflow and open-source components to reduce ad hoc integrations.

Where alternatives often have advantages

Managed cloud platforms (SageMaker, Vertex, Databricks) may be preferable when buyers prioritize reduced cluster operations and accept cloud provider lock-in.
Enterprise Kubernetes platforms (for example, OpenShift AI) may be more compelling when an organization standardizes on a single supported enterprise Kubernetes stack and prefers a single commercial vendor for support and subscription.

	Arrikto (EKF / KFaaS)	AWS SageMaker	Google Vertex AI
Starting Price	KFaaS: public usage-based rates for running and stopped deployments were listed; EKF: not public.	Usage-based; published by AWS but complex across features.	Usage-based; published by Google Cloud; varies by service.
Recommended use case	Teams committed to Kubernetes/Kubeflow; hybrid/on-prem deployments; organizations that want an open-source-based platform packaged for enterprise use.	AWS-native teams seeking a managed ML experience with integrations across AWS services.	GCP-native teams seeking managed MLOps components integrated with Google Cloud infrastructure.
Key strength	Kubeflow-centric portability; GitOps-oriented reproducibility tooling; Rok data layer and Kale for notebook-to-pipeline workflows.	Broad managed feature set and deep AWS integrations.	Integrated GCP ML services and managed MLOps components tied to Google Cloud.
Key weakness	Limited public third-party review evidence; continued operational responsibility around Kubernetes; EKF pricing not transparent.	Potential vendor lock-in and complex cost management.	Potential vendor lock-in and service sprawl; optimal experience typically within GCP.