Skip to main content
Featured

Welcome to the CDF DataOps Practitioner community! Let's Get Started.

  • September 17, 2025
  • 3 replies
  • 61 views

Peter  Arwanitis
Practitioner ⭐️⭐️⭐️
Forum|alt.badge.img

Hello everyone, and a huge welcome to the new CDF DataOps Practitioner community!
We are incredibly excited to launch this space, which we envision as the central meeting point for the technical experts who are hands-on operating Cognite Data Fusion every single day. This community is built for you, and its success will be driven by you.
Our mission is simple: To create a practitioner-driven space for sharing expertise, best practices, and reusable solutions for operating Cognite Data Fusion at scale. This is where we can connect to solve common operational challenges collaboratively, foster a culture of continuous improvement, and accelerate the delivery of reliable, high-quality industrial data.
Who is This Community For?
If you are responsible for the operational health, efficiency, and governance of one or more CDF projects, you are in the right place. This community is for the CDF DataOps Practitioners, the CDF System Administrators, the Platform Owners, and anyone whose role involves keeping the data flowing reliably from source to solution. You're the ones managing configurations as code, monitoring pipelines, and ensuring the platform is secure and performant.
What We'll Talk About
This is your forum to discuss the real-world challenges and solutions related to managing CDF. We encourage you to start and join conversations on topics like:

  • DevOps & The CDF Toolkit: Best practices for versioning configurations with YAML and Git, building robust CI/CD pipelines for deployment, and scripting complex administrative tasks.
  • Data Modeling & Schemas: Practical tips for creating, maintaining, and evolving scalable data models in a production environment.
  • Monitoring & Alerting: How are you monitoring the health of your data pipelines and transformations? What tools and techniques are you using for effective alerting when things go wrong?
  • Governance & Access Control: Strategies for managing datasets, groups, and permissions in complex enterprise environments. How do you ensure the right people have access to the right data?
  • Documentation & User Support: Share your best practices for documenting your CDF project and handling support requests from your internal end-users.

How We Collaborate: A Few Ground Rules
To ensure this remains a valuable and collaborative space, we ask everyone to embrace a few key principles:

  1. Peer-to-Peer First: This community thrives on shared knowledge. If you see a question you can answer, please jump in!
  2. Share Generously: Your solutions, scripts, and lessons learned are incredibly valuable. Don't hesitate to share what has worked for you.
  3. This is NOT Official Support: For urgent, production-down issues, please continue to use the official Cognite support channels. This hub is for knowledge sharing and non-critical problem-solving.

Cognite's Role
You will see Cogniters participating in the discussions here. Our role is primarily to listen, learn, and facilitate. We are here to understand your challenges, connect your valuable feedback directly to our product teams, and help ensure this community is a productive and positive environment for everyone.
Let's Get the Conversation Started!
To kick things off, we'd love to learn more about you. Please reply to this thread and tell us:

  1. Introduce yourself! What's your role, and what industry are you working in?
  2. What's the most interesting operational challenge you've solved with the CDF Toolkit recently?
  3. When it comes to monitoring your CDF pipelines, what's your biggest blind spot or wish-list item?

Thank you for being here. Let's build this knowledge base together!

3 replies

Marius Sørenes
Committed ⭐️⭐️
  • Committed ⭐️⭐️
  • September 24, 2025

Hello!

My name is Marius Sørenes, and i am part of the Equinor CDF DataOps team as a Data Engineer :)

I work mostly with governance, security modeling and access control related to CDF.

 

What's the most interesting operational challenge you've solved with the CDF Toolkit recently?

We recently did a promotion of a solution from dev to test. This involved creating several groups and spaces in cognite-toolkit. 

 

When it comes to monitoring your CDF pipelines, what's your biggest blind spot or wish-list item?

In Equinor we have developed a monitoring tool enabling us to monitor extraction pipelines and transformations across several project in a nice overview. This would be nice if was implemented into cdf. 


Peter  Arwanitis
Practitioner ⭐️⭐️⭐️
Forum|alt.badge.img
  • Author
  • Practitioner ⭐️⭐️⭐️
  • September 24, 2025

Hello everyone,

My name is Peter Arwanitis, and I'm a Solution Architect and Technical Account Manager at Cognite, based in the Munich area in Germany.

While my current customers are in the Oil & Gas sector, my background is deeply rooted in decision-support systems, interoperability, semantic modeling, and automation (I've been a Python enthusiast since 1999!). That passion for building robust, automated systems is a big reason why I'm so excited to kick off this community with you.

We started this hub, alongside the CDF DataOps teams at Aker BP and Equinor, because we believe that the best solutions come from sharing our collective experience. While Cognite provides the platform, the most powerful, reusable templates and best practices will be built by the practitioners who use it every day. We are here to learn from you and improve together.

On a personal note, one of my long-standing "Cognite hobbies" has been tackling the complexities of CDF authorization and group configurations. It's a challenge that has only grown over the last five years, and I've spent countless hours exploring best practices and automation strategies.

Some of you might remember an open-source community project I maintained called bootstrap-cli. Although it's deprecated now, the spirit of that project—creating a "template approach" to enforce maintainability, governance, and reusability—is something I'm passionate about continuing with the CDF Toolkit. I'm always experimenting with how we can better apply core DevOps principles to our work in CDF.

To answer one of the questions from my welcome thread, my biggest blind spot—and top wishlist item—is a way to navigate all CDF configurations more intuitively. I often think of it as a "Google Maps for CDF," where we could:

  • Zoom in/out to move from the high-level picture to the granular details.

  • Drill down into "points of interest" to see metadata and links to CDF resources or external systems like Entra ID.

  • See "traffic jams" through pipeline metrics and color-coded alerts.

  • Get "directions" to understand how different issues and components are connected.

You could call it a CMDB (configuration management database) on steroids!

I'm truly looking forward to discussing these kinds of challenges with all of you. Please jump in with questions, and more importantly, share your own insights, experiences, and solutions. Let's build this knowledge base together.

Peter
(=PA=)
 

Hello everyone!

My name is Daniel Hansen, I work at Capgemini and i am part of the Equinor CDF DataOps team as a Data Engineer

I work mostly with governance, CDF Toolkit and CDF access control

 

What's the most interesting operational challenge you've solved with the CDF Toolkit recently?

Started promoting a solution from dev to test. This required creation of new groups, spaces, datasets, raw DBs, and accesses 

 

When it comes to monitoring your CDF pipelines, what's your biggest blind spot or wish-list item?

Mainly, I look at failing pipelines through the monitoring tool that was developed earlier, and look at the logs, but I rarely action upon this. That is my biggest blind spot, and I would wish I got more insights into how to rectify them