Staff Eng on the detection team.

— Lead the technical architecture and design of Lever’s system configuration team of 8 engineers, focusing over the last year on the global data compliance system.

— Ensured consistent delivery under stringent contractually obligated deadlines, while mitigating risks and coordinating needs between legal, product and engineering stakeholders.

— Built systems at scale which managed legally required discrete data requirements for many millions of documents.

— Built prototypes, wrote code, reviewed PRs, paired with other engineers, and generally pushed the project forward whenever and however needed.

— Focused on designing a system with a clear and sensible api that was both extensible as the product grows and was generally consumable by other engineering teams at the company.

— Designed, architected, built, and deployed a modern full stack application allowing podcasters to upload audio which is converted to text using NLP. Users can then edit the text and the audio is edited to match automatically.

— Hired & Led a team of 4 developers, through project and technical planning to code and deployment using a combination of pair programming, agile methodologies, test driven development, and regular feedback loops.

— Architected and developed the Database Model for the application.

— Led the development of the frontend React application and its complex UI which allows a text editor to represent audio in a contextual and meaningful way.

— Led the development of the Django REST API, including Dockerization, management of asynchronous task management, and data scaling issues.

— Began the design and implementation of our central hospital deployment product as a set of microservices.

— Led the development and extension of our medical data-collection web application.

— Designed, architected, built and deployed an Electron app to visualize deep learning models for clients.

— Extended and improved the company’s primary medical data collection web application to offer better performance and reliability for doctors using it to do image labeling. This resulted in a greater value per hour of labeling.

— Optimized & restructured database to improve SQL query performance by as much as 60x, bringing queries from minutes to milliseconds.

— Built integration libraries to ingest medical data and export in formats for deep learning models.