My Work

SWE Backend Intern @Rebel

Specialize in backend development with Python on GCP, pioneering scalable and reliable cloud-based solutions for the music industry.

  • Python
  • Django
  • Javascript
  • SQL
Project 1

Tech Fellow @CodePath

Empower aspiring developers through mentorship and hands-on coding workshops, leading sessions on advanced programming concepts and industry best practices.

  • Python
  • Latex
  • Node
  • MongoDB
Project 1

Full stack Engineer @TGSE, Inc

Spearhead automated email list process, reducing manual work by 30% and increasing efficiency in candidate sign-up, advocate invitation, responses, and updates.

  • wix_code
  • HTML
  • Javascript
  • css
Project 3

Projects

Screenshot of web app
NLP application

Applying Contrastiver Learning to Idiom Paraphrasing

Accomplishments
  • Tools: ChatGPT API, Bart, Pandas, Numpy, Pytorch, GCP, Data Augmentation Python
  • Engineered an advanced algorithm utilizing a pre-trained encoder-decoder language model, specifically BART, to generate literal equivalents of idiomatic sentences. Achieved superior comprehension of idioms through contrastive learning and fine-tuning of the encoded latent space, resulting in an optimized model with a 60% reduction in processing time.
  • Employed ChatGPT API to enhance data in MAGPIE and complement a small dataset, aligning their formats and showcasing Model-p&m's exceptional transferability. The model consistently outperformed, with significant performance boosts post-merging MWE and augmented MAGPIE, demonstrating effective data augmentation.

About Me

A computational linguistics graduate with a Master's from the University of Rochester, combining expertise in NLP, machine learning, and full-stack development through academic projects like BERT fine-tuning and professional experience as an AI educator and software engineer at startups.

My Resume