UI / UX Design
AI ASSISTANT FOR DEVOPS ENGINEERS
Designed a scalable self-service onboarding experience for an AI assistant, simplifying a complex technical setup while ensuring critical configuration data was captured correctly.
Year :
2025
Industry :
Developer Tools, AI
Client :
Swisscom
Project Duration :
1 year

THE PRODUCT
DevOps Buddy is an AI-powered assistant designed for DevOps engineers working on active projects. By connecting to a project’s codebase and configuration, it allows engineers to ask questions, explore existing code and generate solutions based on the specific project context. The aim is to reduce context switching and help teams troubleshoot and deliver faster.



My role & Focus
I was the sole designer responsible for DevOps Buddy, owning the UX and UI across the product.
This case study focuses on one particularly significant part of my work: redesigning the onboarding process. The goal was to turn a manual, code-based setup into a guided self-service experience that could scale from individual projects to entire applications made up of multiple projects.

THE Challenge
Our discovery showed that low adoption was not caused by a lack of interest. Several teams had already requested access, but their projects were still waiting to be onboarded because every setup had to be completed manually by a single technical administrator. The process also required several critical pieces of information, and entering something incorrectly could affect the entire project configuration. I worked closely with the DevOps engineers to understand these dependencies, identify potential points of failure and translate the technical requirements into a clear and reliable flow.
THE outcome
The redesigned experience allowed technical administrators to onboard projects independently through a guided interface, removing the dependency on a single person. We later extended the same model to support applications containing multiple projects, creating a more scalable foundation for the product.
Onboarding time was reduced from around 2.5 hours to 10–20 minutes, while the adoption rate increased by 21%.

To know more about this poject, contact me at hello@costanzacorsaro.com
To know more about this poject, contact me at hello@costanzacorsaro.com
To know more about this poject, contact me at hello@costanzacorsaro.com
More Projects
UI / UX Design
AI ASSISTANT FOR DEVOPS ENGINEERS
Designed a scalable self-service onboarding experience for an AI assistant, simplifying a complex technical setup while ensuring critical configuration data was captured correctly.
Year :
2025
Industry :
Developer Tools, AI
Client :
Swisscom
Project Duration :
1 year

THE PRODUCT
DevOps Buddy is an AI-powered assistant designed for DevOps engineers working on active projects. By connecting to a project’s codebase and configuration, it allows engineers to ask questions, explore existing code and generate solutions based on the specific project context. The aim is to reduce context switching and help teams troubleshoot and deliver faster.



My role & Focus
I was the sole designer responsible for DevOps Buddy, owning the UX and UI across the product.
This case study focuses on one particularly significant part of my work: redesigning the onboarding process. The goal was to turn a manual, code-based setup into a guided self-service experience that could scale from individual projects to entire applications made up of multiple projects.

THE Challenge
Our discovery showed that low adoption was not caused by a lack of interest. Several teams had already requested access, but their projects were still waiting to be onboarded because every setup had to be completed manually by a single technical administrator. The process also required several critical pieces of information, and entering something incorrectly could affect the entire project configuration. I worked closely with the DevOps engineers to understand these dependencies, identify potential points of failure and translate the technical requirements into a clear and reliable flow.
THE outcome
The redesigned experience allowed technical administrators to onboard projects independently through a guided interface, removing the dependency on a single person. We later extended the same model to support applications containing multiple projects, creating a more scalable foundation for the product.
Onboarding time was reduced from around 2.5 hours to 10–20 minutes, while the adoption rate increased by 21%.

To know more about this poject, contact me at hello@costanzacorsaro.com
To know more about this poject, contact me at hello@costanzacorsaro.com
To know more about this poject, contact me at hello@costanzacorsaro.com
More Projects
UI / UX Design
AI ASSISTANT FOR DEVOPS ENGINEERS
Designed a scalable self-service onboarding experience for an AI assistant, simplifying a complex technical setup while ensuring critical configuration data was captured correctly.
Year :
2025
Industry :
Developer Tools, AI
Client :
Swisscom
Project Duration :
1 year

THE PRODUCT
DevOps Buddy is an AI-powered assistant designed for DevOps engineers working on active projects. By connecting to a project’s codebase and configuration, it allows engineers to ask questions, explore existing code and generate solutions based on the specific project context. The aim is to reduce context switching and help teams troubleshoot and deliver faster.



My role & Focus
I was the sole designer responsible for DevOps Buddy, owning the UX and UI across the product.
This case study focuses on one particularly significant part of my work: redesigning the onboarding process. The goal was to turn a manual, code-based setup into a guided self-service experience that could scale from individual projects to entire applications made up of multiple projects.

THE Challenge
Our discovery showed that low adoption was not caused by a lack of interest. Several teams had already requested access, but their projects were still waiting to be onboarded because every setup had to be completed manually by a single technical administrator. The process also required several critical pieces of information, and entering something incorrectly could affect the entire project configuration. I worked closely with the DevOps engineers to understand these dependencies, identify potential points of failure and translate the technical requirements into a clear and reliable flow.
THE outcome
The redesigned experience allowed technical administrators to onboard projects independently through a guided interface, removing the dependency on a single person. We later extended the same model to support applications containing multiple projects, creating a more scalable foundation for the product.
Onboarding time was reduced from around 2.5 hours to 10–20 minutes, while the adoption rate increased by 21%.

To know more about this poject, contact me at hello@costanzacorsaro.com
To know more about this poject, contact me at hello@costanzacorsaro.com
To know more about this poject, contact me at hello@costanzacorsaro.com



