Notable Projects
WarcParser - Dataset Curation Source
Sep 2022 • 12 min • LLMs
Python system to download WARC files 5x faster (20 mins to 5 mins) using asyncio and enabled scalable multimodal dataset creation from records via concurrent multithreaded processing.
Vyakaran - Grammarly Clone for Hindi Source
May 2021 • 30 min • Natural Language Processing
An end-to-end grammar-checking application for the Hindi language using the Encoder Decoder Transformer model to identify errors (67% accuracy), React front-end with editing capabilities expanding language technology access for 600M+ native speakers.
Bokehlicious - Portrait Mode Effect Source
Dec 2022 • 12 min • Computer Vision
A deep learning method for creating shallow depth of field in images with an accuracy of 0.993 deploying U-Net architecture and image-processing techniques on the NYU dataset.
Volunteering
Sep 15, 2023 • 2 min • Non-profit
I intern as a data scientist to empower nonprofits at Changing The Present, the Amazon of the non-profit world. My skills now tell a story of impact, using numbers to change lives for the better.
✍🏼 Scribing a new chapter for others
May 08, 2019 • 3 min • Life
My volunteer pen empowered a disabled student to pass their exam and write their own story of independence. Small acts of service authored a new chapter in my life of purpose and meaning.