Worked extensively on creating single source of truth for data, using Retool on the frontend with Typescript monorepos on the backend and Prisma ORM for postgres database.

· Added data modeling and logic to allow for highly granular tracking of rewards, spend, and payments. Added Prisma models to create postgres tables and Typescript services to interface with database. Refactored outdated Typescript code for rewards to make it more encapsulated and scalable. Rewrote payments service to use new data model, improve code structure, encapsulation, retryability in Temporal worfkflows, and scalability, and support an easier user experience.

· Created Vault Liquidity and Projected Spend dashboard in Retool for finance team. Calculated and surfaced projected maximum spend for each currency, so finance team knew when and by how much to refill vaults.

· Created, supported and updated tool in Retool for users to create and edit budgets and compare quarterly and monthly budgeted amounts, actual spend and revenue. Used Retool RPC to interact with the Typescript repo on the backend.

· Used AI tools like Cursor and Claude to expedite development.

· As part of effort to develop AI automatic quest generation, worked on ensuring data integrity and formatting for input data.

Part of the winning team of Flipside’s AI Hackathon – created tool to automatically generate SQL queries from user input.

Rewrite Team

· Designed and developed multiple greenfield microservices using Java, Maven, protobuf, gRPC, Kafka, Lima, Postgres, JDBC, Vue, Typescript, Angular, and JavaScript as part of a large-scale rewrite of legacy services. Personally worked on various services and components across a full stack but primarily developed the Alerts Notifier Service, a microservice that processed events and sent notifications of the specified types to users.

· Built a large portion of the Postgres database storing the data used in these alerting services.

Ticketing Team

· Worked with internationally distributed team to maintain and add new functionality to a legacy application for order management and order data. Required extensive code archeology with Java legacy code.

Central Destinations

· Developed a web tool using Typescript and Vue (with Java backend) for proprietary trading data system.

Client Team

· Added various new functionalities as needed to Portware EMS (the eventual subject of the Rewrite Team’s work) using Java and Maven.

· Debugged and resolved time-sensitive tickets for external clients.

Design and develop R code implementing Monte Carlo simulation of historical performance of financial portfolios constructed using machine learning to establish the probability of observing model-driven portfolio returns in a distribution of returns generated via random allocation, and integrating this with other software components in Microsoft Azure hosted platform.