Orange Monochrome Perfume Arrangement with Tropical Shadows
Orange Monochrome Perfume Arrangement with Tropical Shadows
Orange Monochrome Perfume Arrangement with Tropical Shadows

TL;DR

Goal Explore how Generative AI could enhance IT service management by automating incident documentation, note generation, and knowledge reuse across teams.

My Role: Product Designer working under with Design lead Betty and responsible for experience flow design, and prototype building in collaboration with PMs, and engineers.


Problem & User Story

IT service agents were struggling to document incidents efficiently. Each case required pulling information from multiple communication sources, notes, and related records. The process was repetitive, fragmented, and error-prone.

🗣️ “By the time I finish documenting everything, another incident is already waiting.” – IT Service Agent

Existing AI summarization features lacked trust because users often found inaccuracies or missing context. Our goal was to design an AI workflow that could generate reliable summaries and suggestions while keeping human control in the loop.


Where Complexity Lies

  • AI-generated summaries lost accuracy due to inconsistent input data and scattered information sources.

  • Agents hesitated to rely on automation without transparency or validation.

  • The solution needed to integrate seamlessly into existing workflows without disrupting productivity.

💬 Our real challenge was not only building automation but designing for accuracy, trust, and adoption.



Improved user flow

TL;DR

Goal Explore how Generative AI could enhance IT service management by automating incident documentation, note generation, and knowledge reuse across teams.

My Role: Product Designer working under with Design lead Betty and responsible for experience flow design, and prototype building in collaboration with PMs, and engineers.


Problem & User Story

IT service agents were struggling to document incidents efficiently. Each case required pulling information from multiple communication sources, notes, and related records. The process was repetitive, fragmented, and error-prone.

🗣️ “By the time I finish documenting everything, another incident is already waiting.” – IT Service Agent

Existing AI summarization features lacked trust because users often found inaccuracies or missing context. Our goal was to design an AI workflow that could generate reliable summaries and suggestions while keeping human control in the loop.


Where Complexity Lies

  • AI-generated summaries lost accuracy due to inconsistent input data and scattered information sources.

  • Agents hesitated to rely on automation without transparency or validation.

  • The solution needed to integrate seamlessly into existing workflows without disrupting productivity.

💬 Our real challenge was not only building automation but designing for accuracy, trust, and adoption.



Improved user flow

How I approached it

Research and Discovery

We began by reviewing user pain points through existing field research and on-site visits. Agents valued speed but demanded accuracy and accountability. We also mapped the workflow using the 5Ws and 1H framework to understand context and timing in documentation tasks.

  • Key findings:

    • Agents lose time collecting data from multiple tools.

    • AI outputs are only as good as the quality of the input.

    • Users trust AI more when they can review, edit, or verify the results easily.


Design Sprint

We ran a 5-day design sprint with cross-functional partners from AI Research, PM, and UX Design.
Each day focused on one phase of the process — from defining the problem to testing a working prototype.
My role was to co-facilitate ideation, synthesize insights, and lead prototyping.

Sprint outcomes:

  • Defined core problem: agents struggle to document incident steps due to fragmented sources and manual processes.

  • Generated 7+ feature directions, later merged into 4 key AI-assisted flows:

    • Incident summary generation

    • Resolution note generation

    • Field prediction and auto-fill

    • Solution suggestion based on similar past cases



Sprint Miro Screenshot

Prototype Design

We designed an end-to-end workflow inside the ServiceNow workspace:

  • When an incident is opened, AI automatically summarizes it from chat logs and system notes.

  • As the agent resolves the issue, AI generates structured notes and predicts missing details.

  • Finally, the system suggests similar resolved incidents to accelerate troubleshooting.


Results

  • Documenting incidents became significantly faster and more consistent.

  • Agents reported higher trust and confidence in AI summaries.


Results

The big learning! Garbage in equals garbage out! 

A key takeaway from study was the critical importance of input data quality for our AI tools. For AI to assist agents effectively, each incident must be initially summarized with accurate details.

This groundwork enables the AI to provide more precise assistance in return, creating a virtuous cycle of improvement and efficiency. This foundational work is not just about enhancing current processes but is a step towards a smarter, AI-driven operational future.


Next project.
Next project.

Saas

Next project.
Next project.

Saas

Next project.
Next project.

Saas

Prototype Design

We designed an end-to-end workflow inside the ServiceNow workspace:

  • When an incident is opened, AI automatically summarizes it from chat logs and system notes.

  • As the agent resolves the issue, AI generates structured notes and predicts missing details.

  • Finally, the system suggests similar resolved incidents to accelerate troubleshooting.


Results

  • Documenting incidents became significantly faster and more consistent.

  • Agents reported higher trust and confidence in AI summaries.


Results

The big learning! Garbage in equals garbage out! 

A key takeaway from study was the critical importance of input data quality for our AI tools. For AI to assist agents effectively, each incident must be initially summarized with accurate details.

This groundwork enables the AI to provide more precise assistance in return, creating a virtuous cycle of improvement and efficiency. This foundational work is not just about enhancing current processes but is a step towards a smarter, AI-driven operational future.


View all

View all

GenAI Notes Generation

Help IT service agents automatically document and summarize incidents using Generative AI to analyze notes, chats, and related records for faster, more accurate resolution.

Year

2024

Year

2024

Type

Saas

Type

Saas

Client

ServiceNow

Client

ServiceNow

Timeline

12 weeks

Timeline

12 weeks

View all

View all

GenAI Notes Generation

Help IT service agents automatically document and summarize incidents using Generative AI to analyze notes, chats, and related records for faster, more accurate resolution.

Year

2024

Type

Saas

Client

ServiceNow

Timeline

12 weeks