Microsoft
Driving Microsoft’s Copilot coherence initiative across 6 products, empowering 50+ designers to deliver trusted Copilot experiences for 300+ global enterprise customers.
Challenge
As Microsoft integrated Copilot—its AI assistant powered by OpenAI—across enterprise products like Security, Azure, and M365, design coherence became a critical challenge. Each product team operated with its own priorities, design systems, and pace of innovation. Without alignment, the user experience risked becoming fragmented, eroding trust in the AI and making cross-product workflows confusing and inconsistent.
Approach
I spearheaded the Copilot Coherence work stream within Microsoft Security suites (Defender, Intune, Purview, Entra) and expanded its impact across Azure and M365. This involved close collaboration with design systems including Fluent, Security Fluent extension, and Azure Fluent extension, as well as portal designers, content, research, and responsible AI teams. I established tools, guidance, and self-serve Figma libraries to help product designers contribute to coherence with minimal friction. I hosted regular critiques, set up pattern review mechanisms, and drove alignment on foundational Copilot frameworks—like Copilot auto invocation, portal-to-portal handoff, and fallback behaviors.
Impact
The coherence system empowered designers to move faster while maintaining a high design bar. We delivered a growing library of reusable patterns adopted across multiple organizations, enabling consistency at scale. The work fostered alignment at the C–VP level and shaped a shared vision of “One Copilot”—a unified, trusted Copilot experience regardless of the Microsoft enterprise products in use.
TikTok
Envisioning and optimizing TikTok Shop’s early product detail page to unlock millions in profit growth for 870M users across 12 global markets
Challenge
TikTok Shop launched in the US and UK in late 2023 and quickly expanded to 12 global markets. As a new entrant surrounded by dominant players like Amazon, Walmart etc, the platform faced skepticism from consumers unfamiliar with social commerce. TikTok needed to rapidly build user trust, educate customers on its unique value, and deliver consistent shopping experiences at scale—all while navigating strict political constraints that restricted access to user data or research. As the product design lead for the product detail page, I was responsible for shaping a surface that not only supported confident, informed purchasing decisions but also unlocked TikTok’s core advantage as a video e-commerce platform.
Approach
I led the evolution of the product detail page into a flexible platform layer—serving as a foundation for experimentation, business enablement, and differentiated design. To build trust, we introduced platform badges, brand reviews, shop protection info, and experiences unification across entry points. To capitalize on TikTok’s strengths, we integrated seller videos, creator content, UGC, and livestreams directly into the shopping experience. I also redesigned the page architecture to improve screen efficiency, support new social behaviors, and enable modular testing across content types, price structures, and AIGC elements. This redesign aligned with Gen Z shopping expectations and accelerated iteration cycles, particularly during high-velocity periods like Black Friday. I partnered closely with engineering to structure the page into modular sections that met 2-week development sprint targets.
Impact
The product detail page became TikTok Shop’s highest-performing and most extensible commerce surface—boosting discoverability and conversion, accelerating testing velocity, and supporting a wide range of cross-functional business goals. Our design system empowered teams to deliver features like floating carts, AI-powered reviews and product features, and global trust indicators with minimal coordination overhead. The redesigned experience not only increased shopper engagement and conversion but also positioned TikTok Shop as a competitive, trustworthy alternative in mature markets—anchored in a uniquely video heavy, socially dynamic commerce model.
Convoy
Transforming carrier on-road experience to reduce load execution costs, unlock revenue growth, and buy time in the collapsing post-COVID freight market
Challenge
Convoy’s carriers often found themselves multitasking under high pressure—navigating routes, managing communication, and executing complex load requirements—all without structured in-app guidance or responsive support. These gaps frequently led to load delays, execution errors, and idle time that directly impacted drivers’ earnings. In a competitive freight market where reliability is critical, the lack of operational clarity not only frustrated carriers but also jeopardized Convoy’s ability to meet shipper expectations. As on-time performance faltered, so did shipper trust—threatening Convoy’s service reputation and long-term differentiation in an industry driven by precision and consistency.
Approach
To improve the carrier experience and reduce friction during load execution, I led the design of a unified system that addressed both guidance and support in tandem. I began by conducting in-depth ride-alongs and interviews to understand how carriers managed tasks on the road and where breakdowns occurred. These insights informed the creation of Driver Mode—a contextual, step-by-step interface that provided real-time task visibility, progress tracking, and in-app guidance tailored to the driver's current status. Simultaneously, I worked across support, operations, and design teams to align internal tooling with the on-road experience, ensuring that carriers could access consistent support without leaving their workflows. By streamlining the end-to-end journey—from planning to issue resolution—I helped carriers stay focused, reduce delays, and execute loads with greater clarity and confidence.
Impact
The launch of Driver Mode and the redesigned support workflows had a direct and measurable impact on carrier performance. Drivers were able to complete loads with fewer delays and less confusion, as the guided UI reduced missed steps and clarified next actions. Together, these changes reduced friction during time-critical moments, improved on-time delivery rates, and minimized the financial impact of idle time and failed executions. More importantly, the experience helped rebuild carrier confidence in the platform and reinforced Convoy’s ability to deliver consistent, high-quality service that shippers could depend on.
Amazon
Turning around Amazon’s Buy Again to drive multimillion-dollar growth and deliver one of Amazon’s most profitable reorder experiences
Challenge
Amazon’s Buy Again feature was created to simplify the process of reordering everyday essentials. But in late 2018, a redesign (Buy Again V2) introduced usability issues that led to a failed experiment, missed revenue targets, and growing customer frustration. Customers struggled to find key product details, disliked unexpected card interactions, and often bypassed the Buy Again page entirely—opting to search manually or dig through order history. The challenge was to turn a declining experience into a high-performing, scalable solution that met aggressive revenue goals across 18 global marketplaces.
Approach
I led the redesign of Buy Again from the ground up—beginning with usability research to uncover core friction points. Based on those insights, I improved interaction patterns, restored critical product context, and introduced Buy Again Lite—a new entry point embedded in natural user flows like Search and Your Orders. I collaborated across retail, grocery, and B2B teams to scale reorder experiences and align on long-term vision. In 2020, I spearheaded the launch of Buy Again V4: a fully reimagined mobile experience optimized for one-handed shopping, price comparisons, and responsive product cards.
Impact
The redesigned Buy Again experience delivered significant business impact. Buy Again V3 reversed a negative experiment and drove a 21.9% annual profit increase. Buy Again Lite added another 10.76%, while a custom grocery flow contributed multimillion in new revenue. The next-gen Buy Again experience launched in 2020 and yielded a 15.2% profit increase. The design system and CX frameworks I introduced became foundational to Amazon’s global reorder experience—turning a high-friction surface into one of the most profitable and widely scaled features in Amazon retail.
Amazon Kids+
Redesigning Amazon Kids+ mobile reading to lift subscriptions by 1.5%, elevate customer reviews, and engage young readers across global markets
Challenge
Amazon Kids+ offers thousands of books, videos, and games for children—but the mobile reading experience lagged behind. Parents frequently complained about poor usability, missing features, and unreliable performance, leading to negative reviews and churn. When I joined the team, 25% of app store reviews referenced “books,” with an average rating of 1.9 stars. Kids struggled with the interface, and the internal team lacked a tailored experience for young readers. The challenge was to redesign the mobile reader for kids aged 5–12 to be intuitive, enjoyable, and technically viable—across devices, languages, and legal markets
Approach
I led the end-to-end redesign of the mobile reader experience, starting with core feature prioritization across P0, P1, and P2 phases. I collaborated with PMs, engineers, and legal stakeholders to scope the work, define requirements, and deliver on an aggressive timeline. Drawing inspiration from—but not replicating—the Kindle experience, I created a kid-friendly UI with features like bookmarking, zoom, font scaling, vertical reading, and simplified navigation. I conducted comprehension testing with kids aged 7–12 to validate iconography and flows, iterating based on direct feedback. I also localized the experience for multiple international markets and coordinated closely with engineering to resolve device-specific edge cases, from storage limits to UI clipping on notched displays.
Impact
The redesigned experience led to a measurable lift in customer satisfaction and engagement, with a 1.5% increase in Kids+ subscriptions post-launch. App reviews improved, and core complaints about reading usability were dramatically reduced. The updated UI not only better served kids’ needs but also aligned with Amazon’s broader design system and international standards. The project became a model for how to center kid-specific comprehension and behavior in mobile UX—balancing delight, simplicity, and platform scalability.
“Peixin is a great asset to the team and the company. She has a clear passion for her work, with a strong customer focus and high interpersonal integrity. She is quick with new ideas to improve or reimagine our UX offerings, and often takes ideas a step further to address bigger-picture issues like scale and to drive innovation.”
Brett, PM director
"Peixin as a deeply strategic "big picture" UX designer who finds simplifying solutions to otherwise complex user scenarios. I was fortunate to work with her on a variety project during our time at Convoy, and she definitely is one of the brightest UX designers I've ever worked with. Any company would be incredibly fortunate to bring Peixin onboard to solve their most puzzling user scenarios!"
Mike, Eng director
"Peixin stands out amongst many UX designers I’ve worked with, in her ability to get to the heart of the problem, her holistic approach to understanding the customer journey and how much she cares about the quality of her work and its impact. She will raise the bar in any UX organization."
Dorothy, CTO
"Thoughtful, incredibly organized, and customer-focused. These are a few of the many things I appreciate about Peixin. While working together at Amazon, I got to see her grow, collaborate, and design thoughtful experiences for customer."
