MSc Computer Science student and software developer
Computer Science graduate student at New York University with software development, machine learning, cloud observability, and research experience.
Amazon
June 2025 – August 2025 · Seattle, WA
Interned with the Talent Evaluation team within Amazon’s People Experience and Technology organization, focusing on reliability and observability of internal evaluation services. Built a canary stack that injected synthetic traffic to team-owned services in preproduction environments for continuous monitoring, using AWS Lambda, CloudWatch, internal testing systems, and CI/CD frameworks.
NYU Steinhardt
March 2025 – May 2025 · New York, NY
Worked under NSF-funded research led by Prof. Kayla Desportes on the Creative Computing Cookbook project. Integrated interactive Parsons problems using the js-parsons library and built internal tooling to automate content workflows, streamlining contributor onboarding and updates.
Discover Financial Services
June 2023 – August 2023 · Farnborough, UK
Collaborated within the Business Technology team to enhance the Diners Club International portal. Led development of an automated health-check system using Playwright and Java, reducing manual testing effort for the Digital Payments team by streamlining processes and refactoring health-check systems for critical sub-applications.
MSc, Computer Science
Aug. 2024 – May 2026
Coursework: GPU Programming, Efficient AI, Compiler Construction, Parallel Algorithms, Heuristic Problem Solving.
BSc (Hons), Computer Science
Sep. 2021 – July 2024
Coursework: Natural Language Processing, NLU, Knowledge-Based AI, Distributed Systems.
Developed a machine learning model to predict folded secondary structure from RNA sequences using Python, Pandas, Numpy, Pytorch, and mxfold. Combined Structured SVMs trained via SGD with loss-augmented inference and a dynamic programming algorithm based on Zuker and Stiegler (1981). Achieved an F-score of 0.504.
Designed and implemented a neural network for MNIST image classification from scratch in C. Implemented SGD, AdaGrad, and momentum-based updates, achieving 98.08% test accuracy and a training loss of 0.01109.