I'm developing Transcriblate, an offline-first document translation application for scanned PDFs and images. I work across product conception, requirements, UI/UX, development, and testing.
For Transcriblate, I researched and implemented a pipeline for PDF processing, OCR, text extraction, translation, and document re-rendering. I also developed pixel-level image processing for replacing translated text while preserving document backgrounds.
At Oracle India Private Limited, I worked on the Oracle Field Service Cloud mobile application, including performance improvements and device capabilities such as camera, geolocation, authentication, and image processing. Earlier, at IBM India Private Limited, I developed native iOS applications and contributed to architecture migrations from MVC to MVVM and Core Data.

