Red Hat OpenShift Data Science Reviews & Product Details

The biggest win of Data science professional like us while using Red Hat Open shift is that it allows all engineers and scientist to seamlessly collaborate on project. The software helps us to self containerize enabling smooth development to product level deployment. Review collected by and hosted on G2.com.

Sometimes while running multiple notebooks and GPU models the workload gets strained and require more hardware resource to function. Apart from above pain point RHDS has steep learning curve in understanding of Kubernetes and containerzation domain knowledge making it challenging. Review collected by and hosted on G2.com.

What I like best is that it runs the entire ML workflow on OpenShift itself. Jupyter notebooks come up quickly with ready-made images for TensorFlow and PyTorch, and model serving happens on the same cluster, so moving from experiment to deployment is smooth. It's built on open-source tools, which avoids lock-in, and enterprise security and access control come for free. Review collected by and hosted on G2.com.

The learning curve is steep if you're not already familiar with OpenShift or Kubernetes, and initial setup and cluster configuration can take real effort. It's also resource-heavy and can get expensive for smaller teams. The documentation sometimes lags behind new features, so troubleshooting pipelines or model serving issues can take longer than expected. Review collected by and hosted on G2.com.

What I like most is how it provides a fully supported, centralized environment for machine learning. It abstracts away the complex Kubernetes infrastructure, allowing our data scientists to focus purely on building, training, and deploying models quickly rather than managing the underlying operations. Review collected by and hosted on G2.com.

The initial learning curve can be somewhat steep for data scientists who are unaccustomed to containerized workflows or Kubernetes-native environments. Additionally, monitoring compute and GPU usage requires diligent attention to prevent infrastructure costs from scaling up unexpectedly during prolonged training jobs. Review collected by and hosted on G2.com.

Red Hat OpenShift Data Science excels at integrating open-source AI tooling (Jupyter, TensorFlow, PyTorch) directly into enterprise Kubernetes. The UI simplifies notebook provisioning and resource requests, while its hybrid-cloud integrations allow seamless model training and deployment across on-prem and public cloud clusters. Its performance with automated GPU slicing and workload isolation prevents cluster bottlenecks. Furthermore, Red Hat's enterprise support and solid documentation ensure steady onboarding, delivering strong ROI by centralizing our team's AI lifecycle and governance into a single interface. Review collected by and hosted on G2.com.

The primary drawback is the steep learning curve during onboarding. The UI / UX assumes significant familiarity with OpenShift and Kubernetes concepts, which creates friction for pure data scientists who prefer turnkey cloud notebooks like Google Vertex AI or AWS SageMaker. Third-party integrations with external data stores, custom container images, and Git repositories often require manual administrative configuration and YAML tweaking rather than a smooth self-service flow. When troubleshooting resource-heavy pipelines or GPU performance bottlenecks, diagnostic logs can be complex to parse. Lastly, enterprise licensing costs and underlying cluster resource demands can make the pricing / ROI difficult to justify for smaller teams without dedicated platform engineering support. Review collected by and hosted on G2.com.

The best thing about Red Hat OpenShift is its portability. I can share my work or projects easily with my team, which saves a lot of time and avoids configuration blunders. It’s also very reliable, and it honestly makes me feel like I’m sharing a song or a movie like in the old days, from one mobile to another. I just love it—it’s so simple. Review collected by and hosted on G2.com.

The main thing I don’t like about Red Hat OpenShift is that the built-in pod consume a lot of resources, and sometimes it’s difficult to manage the configuration of pods and other components. This can lead to higher cloud bills. Also, most of the time, to resolve any issues or configuration-related problems, I have to depend entirely on the Red Hat team, unlike other applications where solutions are easily available on the internet. Review collected by and hosted on G2.com.

What I like most about Red Hat Data Science is that it provides secure, browser-based coding environments with preconfigured frameworks like TensorFlow, PyTorch, Kubeflow, etc. This means I can open a Jupyter Notebook environment with TensorFlow and CUDA already set up, instead of spending time configuring everything myself.

It also supports automated data pipelines, so I can handle MLOps from data ingestion through to prediction in a seamless way. On top of that, I can perform Git operations by connecting my GitHub repositories, which makes it easier to manage code and collaborate.

I find it flexible for both training and deployment. I can train a model in a notebook, register it, and then deploy it as an endpoint on the same cluster, scaling resources up or down as needed. Model serving is flexible as well.

Since everything runs on OpenShift, access control and resource limits are handled by the platform rather than by me. Finally, the dashboard UI is clean, and it saves me time because it’s easy to navigate. Review collected by and hosted on G2.com.

As of now, I don’t have anything else to add, but it would be better if it supported larger models as well. Review collected by and hosted on G2.com.

I like that it makes it is easy to import data into and analyze it easier within the platform than in an excel sheet. REDCap also offers built-in reporting and data visualization tools, customizable data collection forms, and the ability to manage and export data in a structured format. I appreciate that it supports data quality through validation rules and required fields, which can help reduce errors and improve consistency. Review collected by and hosted on G2.com.

because I didnt know how to use Python when I started using the platform it was hard to get over that learning curve because I had to ask for things I knew what I wanted but didnt know how to do the coding Review collected by and hosted on G2.com.

Easy-to-use interface for data science workflows

Scalable environment for ML workloads

Strong security and role-based access control

Good collaboration for data science teams

Reliable performance for model training and deployment

Reliable enterprise support from Red Hat

Seamless integration with Red Hat OpenShift and Kubernetes

Good value for enterprise AI/ML workloads

Support model training and deployment Review collected by and hosted on G2.com.

Initial setup can be complex

Higher learning curve for beginners

Enterprise licensing can be expensive Review collected by and hosted on G2.com.

Red Hat OpenShift AI provides a seamless experience for setting up GPU-ready Jupyter environments and deploying models into production. It simplifies the process by offering preinstalled tools and container features, ensuring a hassle-free workflow for data scientists and developers. Review collected by and hosted on G2.com.

Having used Red Hat OpenShift AI for a year, I love how quickly I can spin up GPU-ready Jupyter environments with tools like PyTorch and TensorFlow preinstalled. It saves us huge setup time and makes sharing notebooks with teammates completely hassle-free without version conflicts. Deploying trained models directly into production is smooth and painless thanks to its built-in container features. Review collected by and hosted on G2.com.

In my company, we are very attached to Kubernetes as an infrastructure, so we find it very comfortable to work with OpenShift. Moreover, even if we don't always work in small teams of 2/3 people, OpenShift allows us to collaborate more easily compared to other platforms we have tested. Review collected by and hosted on G2.com.

Let's say the pricing is not exactly convenient and the interface is sometimes not the most comfortable, but overall it's fine. Review collected by and hosted on G2.com.