Early work.

What it was

I joined Betterment as an apprentice under founder Jon Stein and worked as a product-design and front-end intern. I focused on making a new kind of investment platform easier to understand as people added goals and managed multiple accounts. The work included summary dashboards, allocation views, and the first multi-account navigation concepts. I also built TrainBoard, an internal office dashboard that turned recent deposits, withdrawals, and goals into a live split-flap departure board.

What carried forward

Betterment was an early lesson in making financial complexity feel calm and legible. Account hierarchy, goal naming, allocation, and progress all needed to stay visible without making the product feel like a spreadsheet; TrainBoard extended the same idea into an ambient, memorable data display rather than another back-office table. That instinct toward approachable systems still carries through the work today.

What it was

A live football product spanning match discovery, scores, standings, lineups, player statistics, and competition navigation. At Applico, I led the end-to-end UX and UI and worked with the front-end team through App Store delivery.

What carried forward

The work treated a match as the product's organizing object. Live, upcoming, and completed games; leagues; teams; players; tables; and lineups all had to remain connected without turning the experience into a dense statistics utility.

What it was

A mobile product for captioning and sharing photos. At Applico, I led the UX, UI, and visual direction across onboarding, creation, feeds, detailed caption states, sharing, and launch presentation.

What carried forward

The shareable captioned image—not the feed—was the most important output. The interface was refined around making that object quickly, previewing it clearly, and carrying it into other social platforms with enough personality to travel on its own.

What it was

A video-first mobile product explored across account setup, content capture, editing, tagging, publishing, and supporting administrative states. The archive preserves both system-level flows and the visual identity work around the product.

What carried forward

The most useful artifact is the breadth of the system map: the creative interaction was designed alongside authentication, permissions, failure states, data entry, and publishing rather than as an isolated camera concept.

What it was

Product and web explorations for the Magnises membership platform, spanning the black-card proposition, member benefits, event and venue discovery, account surfaces, and a conversational concierge concept.

What carried forward

The strongest throughline was access rather than transaction. The digital product had to make the value of membership visible, then reduce the distance between discovering a benefit and acting on it through booking, concierge, or an event experience.

What it was

A shipped iOS nightlife product built around friend groups, shared plans, conversation, event discovery, and an Uber-powered ride flow. The premise was to make the group—not the venue—the primary object in a night out.

What carried forward

The public repository preserves a 60-commit Objective-C and Swift codebase, Parse and Facebook integrations, and two releases—including a February 2015 build prepared for App Store approval. It is an early example of a pattern that still runs through my work: identity and social participation organized around real-world behavior.

What it was

A focused two-month engagement leading all visual and product design for a New York luxury ecommerce startup. The work connected jewelry discovery, personal style guidance, guided gifting, responsive commerce, and mobile product direction into one premium customer experience.

What carried forward

Swoonery sharpened an approach I still use: start with the emotional reason someone is buying, then make product discovery, guidance, and transaction feel like one coherent experience. Filters, jewelry profiles, hint lists, and gifting flows translated personal taste into useful product structure without making luxury feel distant or difficult to use.

What it was

A winning MHacks IV project that showed where people nearby were having the best time, then made it possible to request an Uber directly to that activity. I worked with the team on the product and visual experience during the hackathon.

What carried forward

The system analyzed local Twitter activity with Python and TextBlob sentiment analysis, reverse-geocoded popular venues, stored results in Parse, and surfaced them in an iOS map and ranked event list. Selecting an event deep-linked into Uber with the route and car type already filled in.