My Professional Journey
Capital One Mobile App (iOS - Android)
As the leader of the login team for the Capital One mobile app, I was entrusted with overseeing one of the most essential components of the platform, used by millions of customers daily. My team was dedicated to delivering a seamless and secure authentication experience, integrating cutting-edge technologies while maintaining high performance and reliability. By refining legacy systems and implementing modern solutions, we ensured that the login process met stringent security standards without compromising user convenience.
Beyond the technical execution, I prioritized fostering a collaborative team environment that encouraged innovation and problem-solving. By combining a people-first leadership approach with technical expertise, I empowered the team to tackle complex challenges and deliver transformative results. Our efforts not only enhanced the user experience but also reinforced trust in Capital One's commitment to secure and intuitive digital banking.
National Mall Hyperspectral Material Analysis
Recently, I performed a Hyperspectral material analysis of US National Mall. Utilizing HYDICE sensor data from 1995, I was able to pull out different materials for various monuments. I then took the wavelengths for various monuments and applied them across the entire National Mall to determine the types of building materials used across the area and how they correlated with the chosen monuments (Washington Monument, Lincoln Memorial, and Capitol Dome). Deeper analysis showed that these three materials looked similar but were completely different. Washington Monument was constructed from East Coast Marble mostly from Maryland. The Lincoln Memorial’s marble came from a Colorado mine. Finally, the Capitol Dome is actually painted Cast Iron. These three distinct materials are used across the Mall and tell a story about the history of the US and construction technology of the 19th and early 20th centuries.
This analysis can be applied to many material and even gasses.
Graph of HSI Wavelengths for Washington Monument, Capitol Dom, and Lincoln Memorial
RGB Image of National Mall from HYDICE Sensor
Generated SAM data from HSI wavelength data targeting the 3 monuments
SAM data overlayed National Mall
National Art Gallery
Location SDKs
Displaying the collection of a sample of raw GPS coordinates from the stock Android Location SDK. Then using geodesic math, I convert them and generate a corrected point to find the user’s actual location. (2019)
An example of the Urban Canyon effect, which the above video is showing software correction for.
Location services are a powerful tool in mobile technology, but balancing accuracy and efficiency is a constant challenge. My work in developing mobile software SDKs tackles this head-on by refining location data, minimizing battery drain, and enhancing precision. Using advanced filtering, anomaly detection, and sensor fusion, my solutions clean up raw GPS signals, ensuring businesses and developers get reliable, optimized location data without excessive processing overhead.
Battery efficiency is critical in mobile applications, where constant GPS polling can quickly deplete power. My SDKs implement energy-saving strategies like adaptive sampling, motion-based triggers, and predictive modeling to reduce GPS usage while maintaining accuracy. By integrating Wi-Fi, Bluetooth, and sensor-based tracking, they provide precise location data without sacrificing battery life.
To further improve accuracy, my SDKs leverage multi-source data fusion, combining filtering, signal analysis, and real-time correction algorithms to refine positional data. These enhancements enable reliable location tracking even in challenging environments like urban canyons and indoor spaces, ensuring that navigation apps, fleet management systems, and geospatial platforms deliver optimal performance.
GPS data collection of a drive around downtown DC. Design of this is for accuracy and battery savings. This is highly accurate data for a downtown region, compared to the firehouse of inaccurate datapoints most location services provide. Part of MapBox Analytics SDK (2019)
Some of the raw signal data taken from GPS signals to help determine if a device is indoors vs outdoors. This was used to trigger different types of location correction and venue detection. (2015)