1.

Are Right a Lot

A. Tell me about a time you did not have enough data to make the
right decision (1)

Situation: During my internship at SoFi, I developed a login history feature to enhance

bank account security. The challenge was making a decision regarding which login data
points to prioritize when identifying potential security threats. As the feature had not yet
been deployed in production, I lacked sufficient real-world data, creating uncertainty
about the best approach.

Action:

Drawing on my past experience with security systems and user behavior analytics, I
hypothesized that unusual geographic locations and unfamiliar device types were strong
indicators of potential security threats.

To confirm this intuition, I met with the security team to discuss common risk indicators
and shared my initial thoughts. They agreed that geographic anomalies and new device
logins were significant factors in detecting suspicious activities and gave additional
insights to further guide my decision.

After the meeting, I then created an initial prototype of the login history feature that
specifically tracked login frequency, device types, and geographic locations. I then rolled
out the feature to a small user segment and monitored the login data over a few days.
By analyzing the collected data, I observed patterns that confirmed my intuition. Logins
from unfamiliar devices and unexpected locations were closely associated with flagged
security incidents. This validation gave me the confidence to prioritize these metrics in
the final implementation.

Result: By leveraging my past experience to form an initial hypothesis and then

systematically collecting and analyzing data to confirm it, I made well-informed
decisions about which login metrics to prioritize. The refined login history feature
significantly improved the detection of security issues. Users appreciated the timely
alerts and increased transparency, which enhanced their trust and satisfaction with the
platform.

B. Tell me about a time you made an error in judgment/ a mistake?

What did you learn? (2)
Situation: The biggest mistake I made was when I was working on an intern project. I was in
charge of creating a demo for a new feature that would correctly recommend products for
customers to buy based on their past purchases. The deadline was very tight, and I needed to
finish it within a week. I paid too much attention to the details, specifically model runtime
performance, because I wanted to make the classification system perfect. However, I did not
notice I wouldn’t have sufficient time to finish all the details.

Action: When the manager pointed this out, I apologized to him and proposed a way to keep
the demo. I explained that it was okay to prioritize getting the demo ready before the deadline
over runtime performance. The customer would not notice the slow inference time because this
was a demo. After I presented the demo, I rewrote the model code to improve the performance
to meet all requirements.

Result: The customers really liked the demo. Although I was able to finish the demo before the
deadline, I learned the importance of focusing on the bigger picture from this experience. While
details are definitely important, I also need to pay attention to the entire project schedule. In a
future project, I made sure to place more focus on the bigger picture and consequently,
deadlines were met.

2. Bias for action

Calculated risk when speed was critical or

Give me an example of when you had to make an

important decision and had to decide between moving
forward or gathering more information. (3)
Situation: During my internship at Pullscription, we were in the final stages of integrating a new
payment processing system for our e-commerce platform. The system was crucial for handling
transactions during an upcoming promotional event. However, the API documentation provided
by the payment gateway was incomplete, missing key details about handling errors and
transaction retries. We faced a critical decision: move forward with the integration despite
incomplete documentation to meet the deadline or delay the launch to gather more
comprehensive information and ensure the implementation was robust.

Action: I took the following steps to mitigate the risks and make an informed decision:

● Assessed Urgency and Impact: I evaluated the importance of launching the payment
processing system on time, as the promotional event was expected to drive high traffic
 and sales. Meeting the deadline was crucial for maximizing customer engagement and
ensuring a great user experience.
● Reviewed Available Information: I analyzed the existing documentation and conducted
tests to understand how the payment system handled typical transactions. I also
identified missing information, particularly around error handling and transaction failures,
which posed potential risks if we proceeded without addressing them.
● Consulted with Stakeholders: I organized a meeting with the development team and
manager to discuss the risks of launching with incomplete documentation. We explored
the potential consequences of transaction failures during the promotion versus the
impact of delaying the launch.
● Made a Decision: In the end, I decided to move forward with the integration using the
available information, but with a phased rollout approach. We launched the core
payment functionality with basic transaction handling in time for the promotion but kept
advanced features (like retries for failed transactions) in development. In case of errors, I
implemented real-time monitoring to notify customers and prompt them to retry their
payments manually.

Result: The payment processing system was successfully launched on schedule, enabling
seamless transactions during the promotional event. The phased approach allowed us to
identify and address minor issues in real-time without disrupting the customer experience.
Post-launch adjustments further optimized the system, demonstrating my ability to balance
timely delivery with risk management and continuous improvement in a high-stakes
environment.

Customer obsession

Above and beyond for customer/Difficult customer (4)

Challenge:

During my previous internship at Soma Reality, I was developing a social media mobile app for
clients to share stories with each other. Early in the development process, I noticed that the user
retention rate had begun to decrease, particularly among users from non-English speaking
countries. Although this issue was not raised by the team, I took the initiative to find the root
cause of the low retention rate, as it was negatively impacting the customer experience.

Action:
Determined to understand the root cause of the declining retention rate, I went beyond simply
analyzing user data and sought direct customer feedback. Through reviews and direct
conversations with users, I discovered that many non-native English speakers were struggling to
navigate and fully engage with the app due to language barriers. Additionally, demographic data
confirmed that a significant portion of our user base consisted of non-English speakers.

I proposed a comprehensive solution to the manager: integrating a language selection feature

into the app, adding support for internationalization. This feature would not only prompt users to
select their preferred language upon first use, but it would also ensure that all content within the
app was customized to their selected language. To demonstrate the importance of this feature, I
presented detailed data on our user demographics and highlighted the significant number of
users who were non-native English speakers.

After gaining the manager’s approval, I collaborated extensively with the development team to
ensure that the backend supported multiple languages and that all app content could be easily
translated. To ensure that the translations were not just accurate but also culturally relevant and
engaging, I coordinated with native speakers from different regions to review and adapt the
content. So instead of direct translations, we had them match local context and cultural
nuances. This extra effort made the app not only accessible but also more engaging and
enjoyable for customers from various cultures, going beyond what customers typically expect
from a language selection feature.

Result:

The addition of the language selection feature not only resolved the cultural barrier experienced
by non-English speaking users but also greatly enhanced their overall experience with the app.
This led to several users leaving positive feedback on our company’s website, praising the new
feature for making the app feel more accessible and enjoyable.

Deliver Results
Time you met tight deadline (5)

Challenge: In my internship, I was tasked with delivering a significant payment

processing feature within 2 weeks. To meet the deadline, we had to prioritize the most
critical aspects of the project and explore ways to simplify the work.

Action: I started by breaking down the large task into smaller, more manageable
components. After consulting with the manager, I learned that the core functionality
couldn't be reduced and that testing would be extensive. To address this, I prioritized
delivering the essential features on time and planned to keep all necessary tests in
place. I decided to postpone the implementation of non-essential features to future
releases, to which the manager agreed.
In addition, I also reused similar code during implementation, which saved several
weeks of development time. I maintained regular communication with the manager to
provide updates on our progress.

Result: As a result of these strategies and sacrifices, we successfully delivered the core
features by the deadline while maintaining high quality. The non-essential features were
postponed to future releases, allowing us to meet our immediate goals without
compromising the project’s overall integrity.

Time you met a tight deadline, despite significant, unforeseen

obstacles/almost missed deadline

Situation: During my internship at Pullscription, I was tasked with developing a critical

feature for the company’s e-commerce website that allowed customers to place orders
and make payments. This feature needed to be ready for an upcoming marketing
campaign, so the timeline was very tight.

Action: Midway through the project, we encountered an unanticipated obstacle—a

third-party API we were relying on for payment processing suddenly became unstable,
causing frequent errors and delays. This issue was outside our control but posed a
significant risk to meeting the deadline.

1. Problem-Solving: I quickly assessed the situation and collaborated with the

team to explore alternative solutions. We decided to implement a temporary
workaround using a different API while continuing to monitor the situation with
the original provider.
2. Prioritization: Due to the time lost troubleshooting the API issue, I had to
reprioritize tasks. I focused on ensuring the core functionalities were fully
operational with the alternative API and postponed less critical enhancements
that could be addressed after the launch.
3. Sacrifices: The main sacrifice was deferring the implementation of some
advanced features and optimizations that were part of the original plan. This was
necessary to ensure that the core functionality was delivered on time and met the
business requirements.

Outcome: Despite the unexpected challenges, the feature was successfully launched in
time for the marketing campaign, leading to increased sales and customer engagement.
The workaround proved effective, and after the launch, we revisited the deferred
features and optimizations, further enhancing the product. The successful delivery
under these challenging conditions demonstrated my ability to adapt to unforeseen
obstacles and still deliver high-quality results under pressure.

Dive Deep
Challenge 1: deployment:

A time when I solved a challenging problem was during my previous internship at SoFi. I
had implemented a login history feature, but the most challenging part was when I
wanted to deploy the website on AWS. I found that the generated website JS build file
was very big, over 900 KB, which made deployment very slow. The deadline was tight,
and I needed to solve the deployment issue promptly so that the features could be
delivered to our customers.

Action: I read lots of source code on GitHub and consulted my teammates to gather
their valuable insights. I also tried many times. After performing a deep dive into the
code, log files, and using debugging tools, I realized the problem was that I used a third
party React component, and it had issues with its code quality. The author of that
component used React as prod instead of dev dependency, leading to a big generated
JS file. I then solved the problem by carefully studying the open source code and
rewriting the component by myself.

Result: In the end, the generated JS file dropped to < 200 KB, which made
deployment much faster. I was able to successfully deploy the feature before the
deadline, and I received numerous positive feedback from the team and manager
regarding my approach. When I saw my feature being used by millions of SoFi users, I
felt very proud of it.

Challenge 2: database optimization (6)

One of the biggest challenges I faced during the login history project was realizing that
the system's performance was lagging as the volume of login records increased.
Initially, everything seemed fine during the development phase with a smaller dataset.
However, as we began testing with real-world data, the queries retrieving login history
data started to slow down significantly. It was challenging because it is hard to
reproduce, as the performance issue only occurred sometimes.
I realized the root cause was the way the database was structured and the lack of
indexing on critical fields. The database design was initially established by a previous
team, and as a result, the queries were performing full table scans, which became
increasingly inefficient as the data grew.

To address this, I conducted a thorough analysis of the query patterns and identified the
fields that were most frequently accessed. I then implemented indexing on these fields
and restructured the data storage to improve query efficiency.

These optimizations drastically improved performance, reducing query times by over

60%. The experience underscored the importance of designing for scalability from the
beginning, especially in systems expected to grow over time.

Earn Trust
Critical feedback (7)

Challenge: During my previous internship at Soma Reality, I was developing a social media
mobile app for clients to share stories with each other. During the early stages of development, I
observed a significant portion of our customers originated from countries whose language was
not English. I realized that this could lead to difficulties in the user experience as they struggled
to navigate and understand the app’s content due to language barriers.

Although no one asked me to do anything about it, I felt responsible for this issue as it was
significantly affecting customer experience, and user friendly experience was one of our
company’s key priorities.

Action:

In order to resolve this issue, I proposed the idea to the manager of integrating a
language selection feature into the app, adding support for internationalization. This
way, the app prompts the users to select their language of choice when they first enter
the app, and the text will be customized to the selected language. In order to
demonstrate the importance of this feature, I presented data showing the demographics
of our user base and highlighted the significant number of users who were non-native
English speakers. As a result, the manager agreed to my proposal.

After gaining the manager’s approval, I collaborated extensively with the development
team to implement the feature and ensure that the backend supported multiple
languages and that all app content could be easily translated. I also worked closely with
the UX/UI designers to implement frontends for the language selector.
Result:

The addition of the language selection feature significantly improved the app's usability
for non-English speaking users. Several users left positive feedback on our company’s
website praising the new feature and the number of users increased by X percent. I
learned the importance of proactively identifying and addressing customer needs and
developing innovative solutions to address those needs.

Helped struggling team member (8):

Challenge: In one of my group projects, I noticed that one of my team members, Alex, was
struggling to complete his assigned tasks, which could affect our entire team’s progress.
Action: I decided to approach Alex privately to offer my assistance.

- I set up a meeting to discuss the things that were blocking him.

- Alex had multiple upcoming exams to study for, so he didn’t have time
- I worked together to come up with a plan to lighten his workload. I suggested that he
study for the exams for now and after it is over, he can write the report. He said he is
very skilled at writing.
- This way, he will still be contributing.

Have Backbone, Disagree and Commit

Tell me about a time you disagreed with team member (9):

Challenge: During our final project for my university’s database course, I noticed our project’s
website had a serious performance issue. When a user types in a query to the search bar, the
results page took over 6 seconds to load. When I saw this, I notified the team on the Slack
channel. We agreed that the performance issue was related to database requests, but our
proposed solutions differed. My colleague suggested redesigning the database schema,
while I proposed introducing caching to avoid the risks associated with major changes
close to our deadline. We all felt our proposals were reasonable.
Action: I scheduled a meeting with the team where we sat down together to discuss
and test our proposed solutions through a zoom call. We spent the time writing a
prototype of both methods and then we tested their performance on the computer and
decided which one worked best. I explained my reasoning to my classmate using the
caching approach in that it is easy to implement and can meet performance
requirements before the tight deadline, and he understood my thinking. And of course, I
took the time to understand his perspective and address his concerns. I explained to
him that although his approach was a great long term solution to address existing
issues with db design, it was more important to prioritize meeting the tight deadline.
Through demoing the clear performance impact of my caching solution and showing
him the metrics, I was able to convince my colleague. This convinced the team to adopt
the caching solution

Pros of my approach: easy to implement, can meet requirements before the deadline
Cons of my approach: there were existing issues with db design, colleague’s approach
was more of a long term solution.

Tell me about a time you felt strongly about something but team decided to go in
a different direction

You tried to convince others of your reasoning, had solid data, but due to
deadlines the team went with another approach. You then commit to the other
approach and support it.

Situation:

At SoFi Technologies, I was working on a project to enhance the bank account security
system with a new feature that would allow users to track login history.

I advocated for developing a custom real-time monitoring system for login attempts. My
goal was to provide users with up-to-date information and improve account security.
However, the team was considering an off-the-shelf solution that was more
straightforward but lacked real-time capabilities.

Action:

I presented my case to the team, highlighting the benefits of the real-time monitoring
system, such as improved user insights and enhanced security. Despite my strong
advocacy, the team decided to proceed with the off-the-shelf solution due to the
project’s tight timeline and the need to deliver quickly.
Recognizing the decision and the team’s constraints, I embraced the “disagree and
commit” principle. I committed fully to the off-the-shelf solution, working diligently to
ensure its successful implementation and integration into the security system. I focused
on making the most of the chosen solution, optimizing its configuration, and addressing
any issues that arose during development.

Result:

The login history feature was successfully integrated into the security system on time.
Although it didn’t have the real-time capabilities I initially wanted, the feature still
provided valuable functionality to users. This experience reinforced the importance of
aligning with team decisions and contributing positively, even when the outcome isn’t
what I originally envisioned.

Insist on Highest Standards

Describe a time you were unsatisfied with the status quo (10).

Situation: At my previous internship at Pullscription, our team was responsible for

developing features for the company’s e-commerce platform. We faced a recurring
issue where our code reviews were consistently delayed, leading to bottlenecks in the
development process and impacting our ability to meet deadlines.

Action: I first gathered feedback from team members to understand the root causes of
the delays. I found that part of the issue was due to a lack of clear guidelines and
inconsistent feedback from reviewers. To address this, I proposed the implementation of
a standardized code review checklist and created a shared document outlining clear
review criteria and deadlines. I also organized a team meeting to discuss these changes
and ensure everyone was on board. To further facilitate the process, I introduced a new
tool to track code review status and reminders.

Result: The implementation of the checklist and tracking tool significantly reduced code
review times. Our team was able to cut down the average review cycle from 5 days to 2
days, which allowed us to meet our deadlines more consistently. This improvement also
enhanced team collaboration and morale, as team members felt the process was more
organized and efficient.

By addressing the inefficiencies in our code review process, I was able to help the team
become more productive and satisfied with our workflow.
Invent and Simplify
Complex problem solved with a simple solution (reused solved challenging
problem from Dive Deep)*

Complex problem: database performance issue as it was hard to reproduce, simple

solution as not many lines of code were changed

One of the biggest challenges I faced during the login history project was realizing that
the system's performance was lagging as the volume of login records increased.
Initially, everything seemed fine during the development phase with a smaller dataset.
However, as we began testing with real-world data, the queries retrieving login history
data started to slow down significantly. It was challenging because it is hard to
reproduce, as the performance issue only occurred sometimes.

I realized the root cause was the way the database was structured and the lack of
indexing on critical fields. The database design was initially established by a previous
team, and as a result, the queries were performing full table scans, which became
increasingly inefficient as the data grew.

To address this, I conducted a thorough analysis of the query patterns and identified the
fields that were most frequently accessed. I then implemented indexing on these fields
and restructured the data storage to improve query efficiency.

These optimizations drastically improved performance, reducing query times by over

60%. The experience underscored the importance of designing for scalability from the
beginning, especially in systems expected to grow over time.

What was the most innovative thing you have done (favorite project)*

Most innovative thing done is implementing a predictive analysis feature to help

customers proactively detect unusual login patterns and secure user account
Situation: During my internship at SoFi, I implemented a login history feature for our
bank account security center. The initial customer request was somewhat vague—they
wanted a way to see their login activities to help keep their account secure, but it wasn’t
clear what specific details they needed or how the feature should function.

I needed to figure out what users actually wanted to build a feature that would not just
meet basic requirements but also go beyond to enhance the overall customer
experience.

Action:

1. Collaborated with the Product Manager:

○ I worked closely with the product manager to better understand the
customer needs behind the vague customer request. I requested from her
specific customer support tickets and feedback other customers had left to
learn more about the specific security concerns users had. These
conversations helped me realize that users needed more than just a list of
logins—they wanted detailed information that would help them detect
unauthorized access as fast as possible.
2. Developed the Feature with the Team:
○ Predictive Analytics: To help the customer proactively anticipate and
identify potential security threats, I integrated predictive analytics into the
feature to detect unusual login patterns and automatically alert users to
these risks.
3. Regular Communication with the Team:
○ I kept the product manager and other team members updated on my
progress. I shared my work regularly to get feedback and ensure that I
was on the right track. This ongoing communication helped me make sure
the final feature aligned with both customer needs and the overall product
vision.

Result: The login history feature not only met the initial customer request but also
exceeded their expectations. Users were impressed with the detailed insights and
especially the predictive analytics that proactively identified potential security threats.
These enhancements made it easier for users to manage their account security and
gave them greater peace of mind. This experience taught me the importance of clear
communication, anticipating customer needs, and proactively seeking out information to
deliver a successful product.

Learn and Be Curious

Tell me about a time you needed a deeper level of subject matter expertise to do
your job well (11).

Ownership
Tell me about a time you took on something significant outside your area of
responsibility (also customer obsession)*

Challenge:

During my previous internship at Soma Reality, I was developing a social media mobile app for
clients to share stories with each other. Early in the development process, I noticed that the user
retention rate had begun to decrease, particularly among users from non-English speaking
countries. Although this issue was not raised by the team, I took the initiative to find the root
cause of the low retention rate, as it was negatively impacting the user experience.

Action:

Determined to understand the root cause of the declining retention rate, I went beyond simply
analyzing user data and sought direct customer feedback. Through reviews and direct
conversations with users, I discovered that many non-native English speakers were struggling to
navigate and fully engage with the app due to language barriers. Additionally, demographic data
confirmed that a significant portion of our user base consisted of non-English speakers.

I proposed a comprehensive solution to the manager: integrating a language selection feature

into the app, adding support for internationalization. This feature would not only prompt users to
select their preferred language upon first use, but it would also ensure that all content within the
app was customized to their selected language. To demonstrate the importance of this feature, I
presented detailed data on our user demographics and highlighted the significant number of
users who were non-native English speakers.

After gaining the manager’s approval, I collaborated extensively with the development team to
ensure that the backend supported multiple languages and that all app content could be easily
translated. To ensure that the translations were not just accurate but also culturally relevant and
engaging, I coordinated with native speakers from different regions to review and adapt the
content. So instead of direct translations, we had them match local context and cultural
nuances. This extra effort made the app not only accessible but also more engaging and
enjoyable for users from various cultures, going beyond what customers typically expect from a
language selection feature.

Result:

The addition of the language selection feature not only resolved the cultural barrier experienced
by non-English speaking users but also greatly enhanced their overall experience with the app.
This led to several users leaving positive feedback on our company’s website, praising the new
feature for making the app feel more accessible and enjoyable.

2. Alternate story regarding Pullscription experience.

Challenge:

During my past internship at Pullscription, I was a vital member of the backend team managing
the billing feature. One day, I attended a company meeting and noticed that the conversion rate
between free and paid users of the company’s website was not as good as expected. I felt
responsible for that space as my team owns the billing feature. Although no one asked me to do
anything about it, I took the initiative to check if there was any technical explanation for the low
conversion rate.

Action:

In order to resolve this issue, I dove deep into the log files storing the customer metrics as well
as user analytics. I realized many crucial revenue generating buttons resulting in monetary
benefits were not being clicked at all. After a careful analysis, I suspected the root cause was UI
design, which lacked clarity and failed to prompt users effectively.

I then scheduled a meeting with frontend team to discuss my findings. I carefully explained my
reasoning and presented detailed data analysis on user paths and button usage, as well as
screenshots of the frontend UI for the billing feature. This convinced the frontend team of the
importance of improving UI in resolving the low conversion rate. Consequently, the frontend
team offered helpful suggestions on improving the frontend design to make it more user friendly.

After the meeting, I presented a proposal to the manager. It outlined the justification for
improving the UI design to resolve the low conversion rate, requesting to do an AB test. After
careful consideration, the manager approved the requests.
Result: The results were excellent. The AB test revealed the conversion rate increased from 5
to 9%, so we deployed the changes. This led to a significant revenue increase for Pullscription.
Besides, the UI improvements led to a much better customer experience, as shown by the
numerous positive feedback we received regarding the new UI.

Think Big
Tell me about a time you saw an opportunity to do something much better than
the initial focus (reused from Customer obsession) *

Situation: During my internship at SoFi, I implemented a login history feature for our
bank account security center. The initial customer request was to see their login
activities to help keep their account secure, but it wasn’t clear what specific details they
needed or how the feature should function.

I needed to figure out what users actually wanted to build a feature that would not just
meet basic requirements but also go beyond to enhance the overall customer
experience.

Action:

4. Collaborated with the Product Manager:

○ I worked closely with the product manager to better understand the
customer needs behind the vague customer request. I requested from her
specific customer support tickets and feedback other customers had left to
learn more about the specific security concerns users had. These
conversations helped me realize that users needed more than just a list of
logins—they wanted detailed information that would help them detect
unauthorized access as fast as possible.
5. Developed the Feature with the Team:
○ Predictive Analytics: To help the customer proactively anticipate and
identify potential security threats, I integrated predictive analytics into the
feature to detect unusual login patterns and automatically alert users to
these risks.
6. Regular Communication with the Team:
○ I kept the product manager and other team members updated on my
progress. I shared my work regularly to get feedback and ensure that I
was on the right track. This ongoing communication helped me make sure
 the final feature aligned with both customer needs and the overall product
vision.

Result: The login history feature not only met the initial customer request but also
exceeded their expectations. Users were impressed with the detailed insights and
especially the predictive analytics that proactively identified potential security threats.
These enhancements made it easier for users to manage their account security and
gave them greater peace of mind. This experience taught me the importance of clear
communication, anticipating customer needs, and proactively seeking out information to
deliver a successful product.