Ever AI is the #2 ranked facial recognition company in the USA. I led the architecture and implementation of the iOS facial recognition SDK, which allows customers to perform facial recognition tasks completely on-device, without sending any data to the server. I worked closely with our machine learning teams to manage turnover of their ML models to the SDK teams, and I worked closely with the other SDK teams (Android, Python, and C++) to ensure consistency between our products. I wrote extensive developer documentation and sample apps to help customers integrate our SDK as quickly and easily as possible.
Before there was Ever AI, there was Ever, a consumer app for iOS, Android, and desktop with hundreds of thousands of monthly active users. Ever helps its customers back up, organize, and relive photos and videos. I was responsible for several major new features on iOS, including a new interactive, animation-heavy onboarding flow, and I was part of a larger team that re-wrote the main news feed in React Native.
Ever is the largest iOS project I’ve worked on, spanning many features implemented using several technologies (including Swift, Objective-C, and React Native), all designed carefully and developed with an eye to ease of use, maintainability, and performance.
TribeHR for iOS
TribeHR for iOS (no longer available on the App Store) was a project created as part of the core TribeHR product, an employee-focused, cloud-based human resources information system. The iOS app was created to increase employees’ engagement in their companies by encouraging them to check in on company news, recognize their peers, look up contact details, and request time off.
We spent a lot of time sketching, prototyping, and revising in order to achieve this goal. We came up with an innovative timeline to aggregate all relevant information in one place, making for a very flat information architecture. Users could scroll backwards in time as usual to find news and other posts, but could also scroll forwards into the future to see what was coming up, such as events and colleagues going on time off.
We prized our thoughtful, collaborative approach to design, and we had a similar approach to quality. We focused on shipping a bug-free app with frequent updates, and this paid off in consistently increasing usage patterns.
During my time at TribeHR I made significant contributions to the core TribeHR web product, which was written in PHP. I was responsible for designing and maintaining the RESTful API, which was used in the iOS app. I was also involved in implementing new features and major architectural changes from time to time.
Concussion Blog Engine
OpenText Tempo Box
OpenText Tempo Box was an enterprise-class Dropbox-like system for sharing files. I worked on many of its client apps: major architectural work on the Android, Mac OS X, and Blackberry apps, plus contributions to the iOS app.