Azure Portal Features
Simplifying Complexity in the Cloud
cloud computing
Design systems
ai features
Background
Due to non-disclosure agreements, ongoing and non-public work within Microsoft cannot be publicly disclosed. For further clarification, I can speak directly to high level, publicly available projects I've contributed to. Please contact for clarity.
AKS Run Command + Copilot
Project Overview: The AKS Run Command project aimed to streamline how users interact with Azure Kubernetes Service (AKS) clusters by enabling command execution directly from the Run Command blade using Copilot. As part of the UX team, I focused on integrating Copilot into this experience to reduce user friction, improve command accuracy, and enhance overall usability for both novice and advanced users.
Problem: Traditional workflows required users to switch contexts, reference documentation, or risk syntax errors. This created a fragmented experience and increased the cognitive load for users performing routine tasks.
Design Goals: Our primary goal was to embed Copilot into the Run Command interface to provide real-time syntax assistance, examples, and intelligent nudges. We aimed to reduce command errors, shorten task completion time, and make the interface more approachable for users unfamiliar with CLI tools. A secondary goal was to explore how this integration could scale across other AKS-related workflows, such as YAML editing and cluster configuration.
My Role: I led the UX design for the Copilot integration within the Run Command blade. This included designing the interaction model, drafting wireframes, and collaborating with engineering to ensure feasibility. I also facilitated design reviews with stakeholders incorporating their feedback into iterative design improvements.
Research & Insights: We conducted a usability study to validate our design hypotheses. Participants were asked to complete common AKS tasks using the new Copilot-enhanced interface. The study revealed that users appreciated the inline guidance and examples, but found the separation between command drafting and execution confusing.
Following the study, I refined the interface to unify the command input and Copilot suggestions into a single, streamlined panel. I also introduced clearer nudges and restructured the layout to reduce visual clutter. These changes were validated through internal reviews and a follow-up heuristic evaluation.
As we progressed through usability testing and internal reviews, I identified several friction points in the initial Copilot interaction model—particularly around how users discovered and applied command suggestions. Based on this feedback, I updated our design patterns to better support progressive disclosure and inline guidance.
Impact & Next Steps: The Copilot-enhanced Run Command experience is now live in the Azure portal.
By rethinking the interaction model and refining the Copilot integration, I helped create a more intuitive and supportive command execution experience. These improvements contributed to a smoother, more empowering workflow that better supports both experienced engineers and those newer to Kubernetes.
Project Overview: The AKS Run Command project aimed to streamline how users interact with Azure Kubernetes Service (AKS) clusters by enabling command execution directly from the Run Command blade using Copilot. As part of the UX team, I focused on integrating Copilot into this experience to reduce user friction, improve command accuracy, and enhance overall usability for both novice and advanced users.
Problem: Traditional workflows required users to switch contexts, reference documentation, or risk syntax errors. This created a fragmented experience and increased the cognitive load for users performing routine tasks.
Design Goals: Our primary goal was to embed Copilot into the Run Command interface to provide real-time syntax assistance, examples, and intelligent nudges. We aimed to reduce command errors, shorten task completion time, and make the interface more approachable for users unfamiliar with CLI tools. A secondary goal was to explore how this integration could scale across other AKS-related workflows, such as YAML editing and cluster configuration.
My Role: I led the UX design for the Copilot integration within the Run Command blade. This included designing the interaction model, drafting wireframes, and collaborating with engineering to ensure feasibility. I also facilitated design reviews with stakeholders incorporating their feedback into iterative design improvements.
Research & Insights: We conducted a usability study to validate our design hypotheses. Participants were asked to complete common AKS tasks using the new Copilot-enhanced interface. The study revealed that users appreciated the inline guidance and examples, but found the separation between command drafting and execution confusing.
Following the study, I refined the interface to unify the command input and Copilot suggestions into a single, streamlined panel. I also introduced clearer nudges and restructured the layout to reduce visual clutter. These changes were validated through internal reviews and a follow-up heuristic evaluation.
As we progressed through usability testing and internal reviews, I identified several friction points in the initial Copilot interaction model—particularly around how users discovered and applied command suggestions. Based on this feedback, I updated our design patterns to better support progressive disclosure and inline guidance.
Impact & Next Steps: The Copilot-enhanced Run Command experience is now live in the Azure portal.
By rethinking the interaction model and refining the Copilot integration, I helped create a more intuitive and supportive command execution experience. These improvements contributed to a smoother, more empowering workflow that better supports both experienced engineers and those newer to Kubernetes.
Project Overview: The AKS Run Command project aimed to streamline how users interact with Azure Kubernetes Service (AKS) clusters by enabling command execution directly from the Run Command blade using Copilot. As part of the UX team, I focused on integrating Copilot into this experience to reduce user friction, improve command accuracy, and enhance overall usability for both novice and advanced users.
Problem: Traditional workflows required users to switch contexts, reference documentation, or risk syntax errors. This created a fragmented experience and increased the cognitive load for users performing routine tasks.
Design Goals: Our primary goal was to embed Copilot into the Run Command interface to provide real-time syntax assistance, examples, and intelligent nudges. We aimed to reduce command errors, shorten task completion time, and make the interface more approachable for users unfamiliar with CLI tools. A secondary goal was to explore how this integration could scale across other AKS-related workflows, such as YAML editing and cluster configuration.
My Role: I led the UX design for the Copilot integration within the Run Command blade. This included designing the interaction model, drafting wireframes, and collaborating with engineering to ensure feasibility. I also facilitated design reviews with stakeholders incorporating their feedback into iterative design improvements.
Research & Insights: We conducted a usability study to validate our design hypotheses. Participants were asked to complete common AKS tasks using the new Copilot-enhanced interface. The study revealed that users appreciated the inline guidance and examples, but found the separation between command drafting and execution confusing.
Following the study, I refined the interface to unify the command input and Copilot suggestions into a single, streamlined panel. I also introduced clearer nudges and restructured the layout to reduce visual clutter. These changes were validated through internal reviews and a follow-up heuristic evaluation.
As we progressed through usability testing and internal reviews, I identified several friction points in the initial Copilot interaction model—particularly around how users discovered and applied command suggestions. Based on this feedback, I updated our design patterns to better support progressive disclosure and inline guidance.
Impact & Next Steps: The Copilot-enhanced Run Command experience is now live in the Azure portal.
By rethinking the interaction model and refining the Copilot integration, I helped create a more intuitive and supportive command execution experience. These improvements contributed to a smoother, more empowering workflow that better supports both experienced engineers and those newer to Kubernetes.
ACR + MAR UX Patterns: Container Registry at Scale
Context: Azure Container Registry (ACR) and Microsoft Artifact Registry (MAR) are foundational services for managing container images. I supported the UX design for a unified experience that helps users manage container artifacts across public and private clouds. Additionally, I have leveraged work from previous designers to continue expanding upon existing patterns within these products.
My Role: I worked on refining the layout and interaction patterns for registry creation, artifact browsing, and simplifying complexity. Further details of product design are restricted under NDA.
Impact: My work helped simplify complex configuration flows and improve the discoverability of key features. It also contributed to a more consistent design language across Azure’s container ecosystem
Context: Azure Container Registry (ACR) and Microsoft Artifact Registry (MAR) are foundational services for managing container images. I supported the UX design for a unified experience that helps users manage container artifacts across public and private clouds. Additionally, I have leveraged work from previous designers to continue expanding upon existing patterns within these products.
My Role: I worked on refining the layout and interaction patterns for registry creation, artifact browsing, and simplifying complexity. Further details of product design are restricted under NDA.
Impact: My work helped simplify complex configuration flows and improve the discoverability of key features. It also contributed to a more consistent design language across Azure’s container ecosystem.
Context: Azure Container Registry (ACR) and Microsoft Artifact Registry (MAR) are foundational services for managing container images. I supported the UX design for a unified experience that helps users manage container artifacts across public and private clouds. Additionally, I have leveraged work from previous designers to continue expanding upon existing patterns within these products.
My Role: I worked on refining the layout and interaction patterns for registry creation, artifact browsing, and simplifying complexity. Further details of product design are restricted under NDA.
Impact: My work helped simplify complex configuration flows and improve the discoverability of key features. It also contributed to a more consistent design language across Azure’s container ecosystem.
Headlamp: Open-Source Kubernetes Monitoring Dashboard
Headlamp is an open-source Kubernetes dashboard maintained by Microsoft and the community. I contributed to UX improvements aimed at increasing usability and adoption.
I've collaborated with engineers and open-source contributors to refine the information architecture and visual hierarchy of the software. I also helped shape the design of new features like plugin discoverability and feature views.
The project strengthened my experience working in open-source environments and designing for developer-centric tools when dealing with very granular needs like pod health within a dashboard view.
Headlamp is an open-source Kubernetes dashboard maintained by Microsoft and the community. I contributed to UX improvements aimed at increasing usability and adoption.
I've collaborated with engineers and open-source contributors to refine the information architecture and visual hierarchy of the software. I also helped shape the design of new features like plugin discoverability and feature views.
The project strengthened my experience working in open-source environments and designing for developer-centric tools when dealing with very granular needs like pod health within a dashboard view.
He also worked independently as a consultant, bridging the gap between typeface creators and users, always championing the needs of both parties. Not only is he the co-founder of the web platforms, Typographica and Fonts In Use, Stephen has also penned a regular column for Print magazine and authored the acclaimed book The Anatomy of Type. In 2017, he became an integral part of the nonprofit library and museum, Letterform Archive, as an Associate Curator and Editorial Director.
Supporting Azure Leadership in Interviewing Candidates (2024)
An additional, valuable experience I took part in during my first year at Microsoft was helping to conduct interviews for filling a user researcher role within the Azure team.
The experience allowed me to get insight into experiencing the interview process from the other side of the table and allowed me to have meaningful discussions with Microsoft leadership on key skill sets and remarkable character traits to focus on when identifying talent.
☀︎ / ☽