Himanshu
Creating with code, driven by passion.
Overview
Social Links
About
Aspiring AI/ML Engineer & Full-stack Developer | IIT Jodhpur (M.Tech CSE) | NIT Hamirpur (B.Tech Math & Comp). I'm a builder at heart, fueled by a passion for creating efficient and scalable solutions. My expertise spans Backend and Full-stack Development (Django, React), underpinned by a strong command of Data Structures & Algorithms. Having contributed to projects at IIT Mandi Catalyst and BetterSpeech, I'm now deeply engrossed in the transformative potential of Generative AI, Speech Processing, and Deep Learning. Let's connect and build the future!
Stack
Experience
BetterSpeech
- Contributed to the development of an appointment scheduling platform on Wix, assisting in the transition from the Acuity system for Betterspeech, a US-based startup.
- Designed a visually engaging and user-friendly UI for the appointment page, enhancing user interaction and satisfaction
- Worked with automation tools like Zapier to test and set up cron jobs for appointment reminders, contributing to the foundation for streamlined communication processes.
IIT Mandi Catalyst
-Worked on a full-stack web application for IIT Mandi Catalyst, automating manual data handling and improving data accessibility with PostgreSQL, Django, and AWS. -Designed custom user interfaces with React JS and Material UI to facilitate data analysis and status tracking for startups and incubators.
Projects(9)
A multilingual speech-to-text and speech-to-speech content summarization system designed for Indian languages. The project leverages advanced ASR (Automatic Speech Recognition) and NLP techniques to transcribe, translate, and summarize spoken content, making information accessible across diverse linguistic backgrounds.
- Developed an end-to-end pipeline for speech transcription and summarization in multiple Indian languages.
- Integrated state-of-the-art ASR models for accurate speech recognition and translation.
- Implemented content summarization using advanced NLP algorithms for concise output.
- Automated dataset generation and preprocessing for scalable model training.
- Designed user-friendly workflows for seamless speech-to-text and speech-to-speech conversion.