Greenfield project: Designing and architecting systems from the ground.

Real-time inference: Optimizing performance for low-latency AI workloads.

Microservices & distributed systems: Ensuring scalability and reliability.

Cross-functional collaboration: Bridging software engineering and AI research.

Infrastructure & deployment: Leveraging cloud-native technologies for efficient model serving.

Reduced logging of our application from ~1Tb per day to 90Gb per day. Saving us $1000's a month in AWS and monitoring costs.

Implemented Elasticsearch for full text search across millions of documents with sub 100ms response times. Ingestion is also "real time". Prior search experiences was taking on average 3seconds.

Optimized many PostgreSQL queries, by improving indexing, analyzing index usage, as well as query structure. An example of this is a query that was taking a minute to run on production was reduced to sub 1s by my efforts.

Go to person for query optimization and endpoint performance optimization.

Implemented the most requested feature for Scribe, which is the changing of the click target color.

Implemented feature flagging using an external vendor. This made deployment of new code substantially less risky and unlocked the ability to do A/B testing organization wide.

Conducted dozens of interviews for new candidates (backend and technical walk through)

Worked on redeveloping API layer

Interviewed Developers

Co-developed the checkout experience. All customers go through this React / Python single page app in order to start a subscription.

Lead migration of email service providers to Braze. Ported over millions of user profiles using Go-lang. Created API integration which keeps user profiles in sync with Braze. (Python, Go)

Refactored backend Stripe integration which handles online payments. Integration handles 100% of payments and refunds.

Architected and implemented a feature that allows users to customize the contents of their box based on inventory, dietary restriction, other user preferences and operational constraints.

Developed an internal tool for fulfillment center employees that both shows what needs to be packed into the box and creates a printable list of products and recipes ideas for customers that are packed into every box shipped. (Python Django)

Wrote a feature where a customer who experiences damaged or missing item can be given credit without the need of a customer care representative. (React, Python)

Created user pages where customers can leave reviews, change delivery preferences, and view more information about products.

Added infrastructure monitoring and logging with DataDog. Allows other developers to debug, monitor performance and receive alerts from the production environment.

Added Cloudfront CDN so all static assets are served faster to end users while significantly reducing requests to our web servers.

Worked with Product and Design to establish a variety of A/B and multivariate tests aimed at improving user experience for customers and increasing acquisition and retention targets.