Mar 2014 – Dec 2017 · 3 yrs 9 mos · San Francisco Bay Area

• Post-Yammer acquisition position. Member of the Infrastructure team responsible for accelerating the rest of engineering. Responsible for multiple open-source initiatives around trying to make large distributed systems more resilient.

Sep 2012 – Mar 2014 · 1 yr 6 mos · San Francisco Bay Area

• Post-Yammer acquisition position. Managed the Infrastructure Services/Tools team which focused on empowering development. Designed and implemented a social peer-access host-based authorizing service. Evangelized the use of a resiliency library focused on service dependencies. Implemented a service to provide real-time metrics of services dependencies and real-time configuration.

Sep 2011 – Sep 2012 · 1 yr · San Francisco Bay Area

• Designed and implemented services to aid in the development of services. Projects include the deployment system for everything behind Yammer.com (50+ services). A measurement, real-time analytics, and vertical profiling system for the main Ruby/Rails application (broken down by controller/action/block). A machine provisioning system which makes use of raw Juniper switches, raw machines, and allows for one-click provisioning.

Jun 2008 – Sep 2011 · 3 yrs 3 mos

• A major designer and developer on an internal scalable, fault-tolerant, distributed, and high available object storage system (storage.adobe.com ~4PB) [C++, Python]
• Core designer and developer of an algorithm to handle node failures. Aimed to quickly restore any lost objects without any loss of data. Improved recovery times by several magnitudes.
• Added functionality to take improve internal processes that could be parallelized. Reduced completion times by several magnitudes.
• Added the ability for the system to support live migration of objects while users were modifying the data.
• Solo effort to write a incident response system that monitors, alerting (SMS/e-mail), graphs, and auto stabilizes object storage clusters. All design, implementation, and testing was solo. It also allows for advanced monitoring techniques (e.g. rate-of-change, graph comparison, multi-datacenter, horizontal, and vertical scaling). [Java, Scala, Lift]

Jun 2007 – Sep 2007 · 3 mos

• Developed a feature for QuickBooks that automated a task for the user which was previously a tedious manual operation. [C++]
• Conducted usability research with actual live customers to satisfy user requests and to reduce technical support calls.
• Initiated and developed a white box testing (WBT) project to target the legacy code of QuickBooks. Presented the WBT to my manager and director who have continued my work.
• Volunteered to engineer a direct deposit feature by a deadline that was very optimistic. Completed the feature on time and with all functionality working.