About Me

Vedant Sanil

I am a Research Programmer/Analyst at Auton Lab, Carnegie Mellon University working under Prof. Artur Dubrawski. My work specifically focuses on the applications of machine learning technologies in time series data, speech, audio and language. This translated to my research work with anomaly detection in time series data, ML based cardiovascular disease diagnosis and designing ML models to solve a variety of problems pertaining to the speech modality. I also take interest in building scalable software systems and their deployment in distributed systems.

My Career

Carnegie Mellon University

Research Programmer under Prof. Artur Dubrawski at Auton Lab, CMU

Aug 2020
Research Programmer/Analyst

ReadMyECG

Data Science intern who worked on time-series analysis of relevant biomedical time series data.

May 2020
Data Science Intern

Carnegie Mellon University

TA for Introduction to Deep Learning (11785) at CMU for Spring 2020.

Jan 2020
Graduate Teaching Assistant

Carnegie Mellon University

I worked under various research labs at CMU on machine learning, signal processing and Multimodal ML.

May 2019
Graduate Research Assistant

Aeronatuical Development Establishment

I worked under the flight sim division at DRDO, specifically on researching and integrating flight controllers on UAVs.

June 2017
FSim Div Intern

SRMAUV

I designed the Electronic systems and subsystems for two iterations of the team's vehicles, Sedna and Alpheus.

Sept 2015
Electronics Domain Lead

My Skills

My Projects

Distributed File System

Built a Distributed File System in Java to store, retrieve and manage files across multiple storage servers. Included critical functionality included Directory tree Integrity, Storage Server registration, File Replication, Concurrent accesses/updates.

...   ...

Distributed Voting System using Blockchain

Created a custom blockchain based voting system using proof-of-work protocol to mine blocks and proof-of-authority protocol to cast votes.

...   ...

Speech-To-Text

Implementation of the Listen-Attend-Spell Architecture for end-to-end speech-to-text conversion system.

...   ...