<a href="http://app.storyd.ai">Storyd.ai</a> is an AI-powered presentation tool that helps users create business presentations using data and storytelling techniques.

I designed new user flows and layouts to create a more efficient and user-friendly experience for the Storyd.ai app redesign project.

I managed the client relationship, holding regular meetings with the team to discuss progress, review designs, and incorporate feedback.

I oversaw the development process, working closely with the development team to ensure accurate implementation of the new design. Additionally, I reviewed the development output, provided feedback as needed, and tested the final product to ensure it met all requirements and delivered a seamless user experience.

Storyd.ai is an AI-powered presentation tool that helps users create business presentations using data and storytelling techniques.

This project was a full redesign of the existing product.

The previous version had two core issues:

Users struggled with clarity. A product that promises structured storytelling ironically felt unstructured in its own UX.

But the product experience didn’t reflect that.

Storyd.ai - Create business storytelling presentations, in seconds

Space is where you as an organization collaborate, manage access, subscriptions, licenses and track your blocks and its details. Manage everything related to Appblocks in one space.

I created wireframes and prototypes to conceptualize and refine the platform's layout and interactions. I designed the user interface (UI) for the AppBlocks app development platform, ensuring a user-friendly and visually appealing experience, and crafted the landing page to effectively communicate AppBlocks' value proposition and features to potential users and visitors. 

Additionally, I developed and maintained a comprehensive design system for the AppBlocks platform, ensuring consistency and efficiency in design implementation. To drive continuous improvement and user satisfaction, I conducted regular user reviews and gathered feedback to iterate on the UI/UX design. I also collaborated closely with engineers to ensure seamless integration of the design into the development process.

TAFFI is an AI-powered agentic commerce platform built to improve how users discover, explore, and purchase fashion online.

Instead of relying on search, filters, and static recommendations, TAFFI introduced intelligent shopping experiences that actively guided users through discovery, styling, and purchase decisions using AI.

Delivered primarily as embedded widgets inside luxury and premium e-commerce stores, the platform helped retailers improve conversion, basket size, retention, and shopper confidence at scale.

TAFFI was founded by a leadership team combining 

deep fashion domain knowledge and advanced technology expertise

, enabling the company to build a strong intersection of 

, validating TAFFI’s AI-first approach to commerce and enabling accelerated product development, experimentation, and enterprise partnerships.

This foundation created an environment of 

rapid iteration, frequent pivots, and continuous product evolution

, owning UX and UI across TAFFI’s evolving product surfaces—from core consumer experiences to AI interaction design.

My responsibilities spanned the full product lifecycle, including concept exploration, high-fidelity UI, interaction design, interactive prototyping, and motion demos used to explain complex AI behaviors. I handled developer handoff, participated in implementation reviews, and conducted hands-on UX QA to ensure design intent translated accurately into production.

I also built and maintained a small-scale design system in Figma, defining reusable components and style tokens that ensured consistency while enabling speed and flexibility as the product evolved.

Beyond the core product, I contributed to TAFFI’s brand and go-to-market execution through Webflow website updates and presentation decks, supporting the company’s transition from a stylist-centric product to a luxury, enterprise-ready platform.

TAFFI operated in a speed-first environment where learning came from real usage rather than long research cycles. The design approach was pragmatic, execution-heavy, and optimized for continuous iteration.

Work typically began with Jira tasks shaped by leadership priorities, live feedback, and product analytics. I explored multiple solution directions early, then moved quickly into high-fidelity design in Figma using real content. Interactive prototypes were essential for communicating complex AI behaviors and aligning with stakeholders.

Design, development, and validation happened in parallel. I worked closely with engineers throughout implementation, conducted real-time UX QA, and refined flows based on production behavior and analytics.

Successful flows became core patterns, while underperforming ones informed the next iteration cycle.

When I joined TAFFI, the product already existed as an early-stage 

. My initial responsibility was not to introduce new features, but to 

understand and stabilize what was already live

I began with a detailed audit of the existing experience, focusing on the home screen, navigation structure, content hierarchy, and micro-interactions. This phase involved identifying friction, improving clarity, and refining interaction details. Several of these improvements were supported by clear before-and-after UI changes, which helped establish a stronger baseline for future iterations.

This early work shaped how I approached the product going forward: prioritizing 

, clarity over complexity, and decisions grounded in real usage rather than assumptions.

In the D2C phase, fashion shoppers could browse products easily, but the experience often failed at the moment of decision-making. Individual items were visible, yet users struggled to understand how those pieces would come together as complete outfits for specific occasions, body types, or personal styles.

Without sufficient context or reassurance, shopping became uncertain. Users hesitated, delayed decisions, or purchased items without confidence in how they would work once delivered. This disconnect between browsing and confident decision-making became the core problem TAFFI set out to solve.

, rather than a conventional marketplace.

Instead of asking users to assemble outfits on their own, the experience centered around 

 designed for real-life situations such as events, travel, or wardrobe upgrades. Users typically came to TAFFI seeking guidance and reassurance, not simply access to products.

The platform allowed users to explore styled outfits, shop individual pieces from those looks, or book one-to-one styling sessions—shifting the experience from transactional browsing to intentional, confidence-driven shopping.

During the D2C phase, I worked across the core consumer experience, designing stylist profile pages that highlighted curated looks and styling approaches, a vertical discovery feed centered on complete outfits, and end-to-end outfit-to-product shopping flows. I also designed the stylist session booking experience, early style preference capture, and personalization logic, while continuously refining onboarding and feature flows as the product evolved.

TAFFI transitioned from a standalone marketplace into an 

, integrated directly into existing e-commerce stores.

This shift moved the business from a D2C-dependent model to a 

, enabling faster scale and clearer value for retailers by meeting customers where they already shopped.

The core design challenge changed fundamentally. The problem was no longer how to attract users to TAFFI, but how to introduce styling guidance and personalization 

—without disrupting trust, performance, or existing conversion flows.

The experience had to feel native, lightweight, and genuinely helpful, while operating within strict technical, branding, and performance constraints.

The product evolved across several dimensions. A dedicated app became a set of embedded Shopify widgets. Human-led styling workflows shifted toward AI-powered assistance. Static curated collections gave way to dynamic outfit generation, and scheduled styling sessions were replaced with instant conversational guidance available at key moments in the shopping journey.

Designed chat-based shopping flows guiding discovery, clarifying intent, and presenting recommendations.

 Improved engagement, session depth, and shopper confidence.

Designed natural language and intent-driven search UX with contextual refinements.

 Faster discovery and improved search satisfaction.

Designed AI-powered outfit composition UX that pairs items across categories into complete, real-time styles, while integrating product-specific Q&A directly into the shopping experience.

 Improved outfit discovery and higher AOV potential.

This pivot introduced an entirely new class of design problems. Widgets needed to feel integrated rather than intrusive. AI interactions had to handle ambiguous fashion intent without overwhelming users. Personalization had to be delivered without heavy onboarding, and trust needed to be built in automated recommendations—all while operating within the performance constraints of live e-commerce environments.

The work described here ran in live production environments across enterprise retailers.

TAFFI’s AI widgets reached millions of shoppers, generated high volumes of real interactions, and supported multiple engagement models including conversational shopping, discovery feeds, multimodal search, and virtual try-on.

These outcomes were driven not by isolated features, but by 

system-level UX coherence across surfaces

 Static screens were insufficient for explaining AI interactions.

 Built interactive prototypes and motion demos.

 Faster development, fewer errors, higher implementation accuracy.

Stakeholder Rejection & Feasibility Constraints

 Some design directions were not technically scalable.

 Close collaboration with CTO and engineers, rapid iteration.

 Higher quality, production-safe design solutions.

 Latency, inference delays, and data constraints.

 Designed progressive disclosure, smart loading UX, fallback flows.

 Smooth, predictable, and trustworthy user experience.

1. Designing in Ambiguity Startups don't have clear requirements or validated assumptions. I learned to make defensible design decisions with incomplete information, then iterate based on real outcomes. 

2. Collaboration Over Handoff Working directly with engineers made me a better designer. I understood feasibility constraints, shipped cleaner work, and caught implementation issues earlier. 

3. Lean Into Constraints No research budget? Use analytics and production behavior. Solo designer? Build reusable systems. Small team? Move fast and learn continuously. 

4. Strategic Product Thinking I didn't just execute designs—I helped shape product direction through design exploration, advocacy for user needs, and strategic UX thinking. 

5. Ship, Learn, Iterate Perfect is the enemy of shipped. I learned to balance quality with velocity, getting real user feedback faster than designing in a vacuum.

TAFFI’s journey—from a D2C styling platform to a B2B AI commerce layer—was shaped by 

timely pivots, pragmatic design decisions, and continuous iteration

Working through this evolution gave me hands-on experience designing across product pivots, AI-driven systems, and real production constraints, while operating in a lean team environment where speed and clarity mattered more than perfection.

Taffi - Designing an Agentic Commerce Platform

Your strong work ethic and dedication to your job are truly appreciated and valued by the team. Keep up the great work!

Chandulal

Appreciation to Navaf for delivering clean and outstanding design work on the Veox.ai project. The execution was well-aligned with the customer's goals, received strong feedback, and played a key role in helping us scale the account. Well done, and thank you.

Storyd.ai - Create business storytelling presentations, in seconds

Navaf Muhammed