Fidelity Investment

UX Research
UX Design
Artificial Intelligence
role
UXD Intern
duration
June - August 2024
disclaimer
The following text provides a reflection on my experiences during the internship. This page does not contain any confidential information from Fidelity nor any project-related imagery.
overview
"The Fidelity Center for Applied Technology®, or FCAT, is a catalyst for breakthrough achievements in research and tech. We assess, test and scale concepts and ideas that advance Fidelity's market leadership and enhance every customer's experience."
In summer 2024, I worked as a UXD intern at Fidelity's FCAT group. I was tasked with researching and designing an education tool that leverages the large language model (LLM). While I am unable to share any details related to this project, this is a high-level overview of my design process!
define

Desirability, Feasibility, and Viability

A primary focus here is validating desirability by aligning with user needs, while concurrently assessing feasibility in terms of technical constraints. This approach ensures that the solutions I develop are both viable and tailored to meet specific demands.
A venn diagram of desirability, feasibility, and viability.
User Research

Interviews

My team and I conducted interviews with users from diverse groups to validate the product's desirability, and we categorized them into a primary persona and a secondary persona. During this process, my team and I took extensive notes and gathered numerous insights. However, we faced challenges in precisely identifying the key user pain points and needs throughout their journey. In response, I created a user journey map to better capture and address these critical areas.

User Journey Map

I developed a user journey map to translate and visualize the findings and insights from our interviews. This detailed map allowed us to delve deeper into our users' experiences, systematically highlighting critical pain points and pinpointing opportunities for enhancement. By mapping out each stage of the user journey, we gained a comprehensive view of the user experience, which was instrumental in guiding our design decisions and improving the overall product strategy.
Image source: https://miro.com/customer-journey-map/
Customer Journey Map

1. Deaf and Hearing-Impaired

Pain points:
  • Limited Access to Non-Verbal Cues: Lack of visual cues.
  • Lack of Real-Time Feedback: Immediate feedback is limited.
  • Background Noise Interference: Noise disrupts comprehension.


Needs:
  • Real-Time Feedback: Visual indicators for success.
  • Visual Communication Support: Real-time captioning, sign language interpretation.
  • Clear Communication: Tools to enhance speech clarity.

2. ADHD

Pain points:
  • Attention Difficulties: Maintaining focus during calls can be challenging.
  • Listening Comprehension: Understanding and retaining information can be tough.
  • Organization Challenges: Keeping track of key points and action items is difficult.


Needs:
  • Distraction Management: Tools to reduce interruptions.
  • Active Listening: Features promoting comprehension.
  • Structured Support: Visual aids and prompts for organizing thoughts.

3. Introverts

Pain points:
  • Phone Call Anxiety: Introverts often feel anxious during phone calls.
  • Preparation Stress: Preparing scripts can be stressful.


Needs:
  • Confidence Boosters: Prompts for call confidence.
  • Assistance with Scripting: Features to ease script creation.
ideation

Competitive Analysis

Although we haven't gotten too much into the design part when I wrapped up my internship, I utilized the early stage research findings and white boarded some ideas to explore what potential design solutions would look like. In this process, I had the opportunity to conduct competitive analysis on existing products, which provided us with crucial insights and highlighted potential areas for innovation.
Image source: Baymard Institute
Competitive analysis example

How do people learn with AI?

One key insight I learned from the competitive analysis was that a lot of the learning tools that leverages AI and LLM fell under a chatbot query modality. Different learning techniques were integrated into chatbot dialogues such as the Socratic Method or the Feynman Technique. However, writing an effective prompt to the AI is hard, and as Jakob Nielsen says, half the population can't do it. On the other hand, because most of the competitors adopted the prompt-based AI user interaction, I felt a bit stuck in imagining what are the other modes of interacting with LLM, and what are feasible.
‍‍Image source: Baymard Institute
Skeleton wireframe with AI chatbot interaction

AI Exploration

I am actively researching and exploring innovative ways to extend the use of AI beyond standard chatbot interactions. My goal is to engage users effectively to meet users where they are regardless of their technical experience with AI and prompt engineering. Additionally, I have created wireframes and mockups to visualize different types of AI-powered personalized learning experiences, which are tailored to meet the unique learning styles and needs of each user. This approach not only enhances user engagement but also optimizes the learning process through tailored educational tools.
https://www.linkedin.com/pulse/prompt-driven-ai-ux-hurts-usability-jakob-nielsen/
An image with text: Articulating ideas in written prose is hard. Most likely, half the population can't do it. This is a usability problem for current prompt-based AI user interfaces.
reflection

Next steps

  • Continue to engage with a variety of users to further narrow down an "innovation sweet spot" that aligns with user desires and technological capabilities.
  • Create additional mockups and wireframes to visualize and explore potential solutions, enhancing the clarity and effectiveness of the design concepts.
  • Focus on improving the usability of the AI features, making them more intuitive and accessible for users with varying levels of technical proficiency.

Takeaways

My UX design internship at Fidelity offered a unique opportunity to immerse myself in the product design process, particularly in conducting extensive user research to validate desirability. Reflecting on this experience, I am deeply appreciative of the complexity involved in harnessing AI to create intuitive and accessible user-centered solutions. This journey has significantly sharpened my skills in creating intuitive, AI-driven solutions, reinforcing the importance of continuous user engagement to pinpoint innovative sweet spots.

Special thanks to: FCAT AIX, FCAT DSAI, FCAT Design, and fellow Fidterns! 🐸

Other projects

Triton Mentor
Vanguard Robo-Advisor Dashboard
ConvoAI
UNC Campus Health Usability Testing
Made by norry 🧋© 2023
Linkedin