projects

Selected work

A closer look at the products and systems I’ve helped design — from early-stage startups to established tech teams

siro

Benchling

verily

google

siro

Benchling

verily

google

siro

Sales Enablement Platform

Siro is an AI-powered sales enablement platform used by businesses to analyze conversations, identify coachable moments, and improve performance across the organization. It helps teams understand what drives successful deals, provides personalized feedback to reps, and scales best practices through data-driven insights.

Redesign

Led a full redesign of all major product surfaces to better reflect Siro’s evolving capabilities

CRM integration

Net new capabilities to integrate with external CRM tools — a key step in landing enterprise clients

Handoff

Helped hire Siro’s founding designer, setting up both the product and team for their next stage of growth

Creativity in enterprise design requires balance. I often explore through spectrums — testing how ideas change as you move toward one extreme or the other.

Enterprise
Consumer

Needs to be intuitive enough to pick up and just get started regardless of your tech savviness

Explorative
Prescriptive

We know what would bring the rep the most value, Siro to exert opinion

Longer session
Shorter session

Reps are using the app in their truck in between visits - not at home/office

sometimes the easist way to make progress is by giving people something to react to

It’s just as important to show more than one path — comparing a thing against itself rarely leads to clarity

A few of the sources that shape my design explorations — from analytics to real user feedback

"That's a game changer, bro. That's a game changer that makes life so much easier."

sales rep manager

49%

increase in usage for a key feature

It's not enough to technically build good feature if users can't find it or integrate it in their workflow

benchling

Studies Platform

A research tool that helps scientists design, run, and analyze studies—connecting data and collaboration across the drug development process.

Enables study managers to generate first-draft reports by leveraging existing study data and documentation

Enables study managers to generate first-draft reports by leveraging existing study data and documentation

ai report generation

Enables study managers to generate first-draft reports by leveraging existing study data and documentation

ai report generation

Gives research teams the flexibility to define structures and guardrails when creating studies

Gives research teams the flexibility to define structures and guardrails when creating studies

schema configuration

Gives research teams the flexibility to define structures and guardrails when creating studies

schema configuration

verily

Care Console (for chronic condition management)

The Onduo Care Console is a healthcare management tool used by health coaches and clinicians (endocrinologists, pharmacists, etc) to bring members simple, personalized, and proactive support to better manage their chronic conditions in their unique life circumstances. The focus is on type 2 diabetes and additional support for hypertension, depression, and sleep apnea. 

Expand capabilities

Enabled clinicians to diagnose members with sleep apnea and conducted physician training

AI for mental health

Created flow for users to measure depression in a more passive, low friction way via voice logging

V1 to V2

Developed a new framework for the care console to support additional chronic conditions

Why do we even need V2? Onduo’s initial focus was only on type 2 diabetes and the types of support we gave members were limited. As more features were tacked on, the product became unwieldy and didn’t fit the needs of the product vision.

we documented how people move around in the product…

… and conducted card sorting exercises to develop a whole new layout

A few snippets of the final design

Note taking is a pretty big part of most clinicians workflows and our product was no different.

Discussion

Discuss with product team what we know vs don’t know

Quant+Qual Data

Usage data, research sessions with current users

Forecasting

How do we make sure this is in line with the overall vision

Brainstorm

Figjam with the product team to gather ideas

A few key takeaways:

  1. it's hard to find specific things • when was the last time I talked about a1c tests with this member?

  2. it's hard to understand quickly • what happened to my members since I'm back from vacation?

  3. it's time consuming • because you take notes on everything you do!

  4. the future is a lot more collaborative • there's another clinician for nearly every condition we add

But why? it boils down to notes being a catch all for a million different things, for every product gap, users would essentially use notes as a way to bandage the issue

Since we used to be part of Google… of course we should add a search bar? But as I dug into it more, i realized this was addressing the symptom but not the actual problem.

It was making it easier to find specific notes but it doesn't address the fact that there's just too much going on in notes to begin with!

the final design for notes ended up having very little to do with… notes. Instead we created a dedicated space for things related to goals + behavior change.

:-)

biggest ratio between engineering cost vs care team satisfaction

this isn't the flashiest project. there's no ai magic here. just digging deep into the user's pain points which ended up being the feature that users were more excited and satisfied with.

100+

pieces of feedback collected

from our v2 pilot launch via roundtable research sessions

Google

Case Management Tool

Cases is the cornerstone of Google’s high-touch support suite, which provides a consistent customer and agent experience across the Ads and consumer-facing Google product base. In 2017, over 60 million interactions were handled by 30k unique agents. Support agents are located in countries all around the world, and the interface is localized to 2 additional languages other than English.

Each designer on the team holds their own unique set of expertise. I was the lead designer for: Product infrastructure, machine learning, accessibility, patterns & components & Cases v1.0 turndown

Hardware support

There’s a long list of rules, policies, and things that can go wrong with hardware.

Created streamlined processes so agents don't need to use discretion.

Suggested Actions

Use machine learning to streamline operational processes and repetitive actions.

Developed a set of patterns to display ML driven suggestions.

Rerouting

We get thousands of new cases per day, sometimes these cases don’t always end up in the right spot. 

Made it fast & easy to reroute using the new method of taxonimization.

6%

Revenue increase

Google Store customers can make purchases without a Google account

1.5hr

decrease in resolution time

email resolution time decreased from 9.38hrs to 7.88hrs

343k

RMA issues resolved each year

streamlined RMA flow that integrates with Payments, Store, Salesforce, etc

From Series A to B, design that keeps up with growth

Fractional UX design for early-stage B2B SaaS startups

From Series A to B, design that keeps up with growth

Fractional UX design for early-stage B2B SaaS startups

From Series A to B, design that keeps up with growth

Fractional UX design for early-stage B2B SaaS startups

© 2026 faith jin. All rights reserved.

© 2026 faith jin. All rights reserved.

© 2026 faith jin. All rights reserved.