I developed PancreaVision, a Flask-based web application that uses a ResNet50 model to classify pancreatic CT images and generate confidence scores.
For PancreaVision, I also preprocessed medical imaging datasets and evaluated model performance using Python.
I developed a GPT-2 conversational system using Hugging Face Transformers, with preprocessing, tokenization, and evaluation pipelines for conversational datasets.
As an LLM Post Training Intern at Ethara AI (Green Rider Technology LLP), I annotated and validated LLM datasets, evaluated model outputs against quality benchmarks, and supported post-training workflows through dataset cleaning and quality assurance.
