Manage a team of data scientists using the Scrum methodology to continually develop and deliver scalable and performant machine learning platforms on-time, communicating results directly to VP and SVP stakeholders.

Propose, design, and implement state-of-the-art machine learning tools to extract business insights at the scale of billions of customer-client interactions per day:

• Causal inference methods measuring and optimizing the impact of digital advertising
• Deep learning to understand the path-to-conversion
• Game theoretic ideas attributing purchases to advertising
• Petabyte-scale, distributed computing

Redesign existing Scala and Python source:

• Reducing overall resource usage by over 50%
• Reducing source lines and complexity by over 70%
• Greatly expanding platform expressiveness and reducing time-to-launch of new products

Organize and curate weekly journal club to keep the department up-to-date on modern techniques and establish shared fundamentals

Developed and implemented modern machine learning algorithms to accelerate and to reduce the cost of materials discovery. Focusing on functional materials like piezoelectrics, combining materials science, geometrical and structural data and DFT simulations, we proposed novel materials for synthesis.

Publication: A Perspective on Materials Informatics: State-of-the-Art and Challenges, 2015

My first exploration into industry, where I worked as a Wet Process Engineering Intern, applying ideas from chemical engineering to supporting production of volatile and non-volatile silicon memory devices.