Mar 2021 – Apr 2025 · 4 yrs 1 mo · Toronto, Ontario, Canada

• As part of the ML Core team, I developed cross-platform machine learning and data analysis solutions to support and improve Workday applications and services.
• Created and deployed a metrics pipeline across Workday applications for analysis of user behaviours, service performance, and implicit user feedback data
• Improved automatic account recommendation for financial applications in Workday using ML models based on semantic analysis.
• Created improved recommender engine for Workday search applications based on knowledge graphs.

Aug 2019 – Feb 2021 · 1 yr 6 mos · Toronto, Ontario, Canada

• Worked with Planning and Optimization teams within Amazon's Fulfillment Technologies; designed and developed state-of-the-art workforce planning tools, using machine learning models, critical for staffing and management of millions of employees across hundreds of Amazon businesses worldwide.

Sep 2015 – Aug 2016 · 11 mos · Seoul, South Korea

• Worked with the Core Artificial Intelligence Lab at the Samsung AI Center in Seoul (formerly known as Samsung Seoul R&D Center)
• Applied machine learning algorithms in vision, scene recognition and navigation of autonomous bots for factory automation. Developed deep learning techniques for computer-vision and image processing applications in smart home appliances.

May 2015 – Aug 2015 · 3 mos · Bengaluru, India

• Worked with the Programming Languages and Tools (PLATO) Group. Extended the angelic verification technique to .NET framework, specifically to C# programs with unconstrained environments.

May 2014 – Jul 2014 · 2 mos · Gyeonggi, South Korea

• Worked with the Intelligent Solutions Team at Samsung Electronics' Software R&D Center.
• Created image tagging program using NLP and supervised learning to associate images from the web with characterizing text.

May 2013 – Jul 2013 · 2 mos · Kaiserslautern, Germany

• Research on verification problems of relaxed memory models for concurrent programs. The goal was to build a reachability checker (based on the EEC algorithm) for programs that run under the Total Store Ordering (TSO) memory model.