Things I've
Worked on
TerraVide
A Python Package to extract and analyze urban environments scanned with LiDAR. I wrote it to help researchers, students and developers break the barrier of working with point cloud data. You can extract features from sensors in realtime and classify approximately 500,000 data points in a matter of seconds
Vistara
An Interactive web app to reduce barrier to entry to using point cloud data. It lets you download any point cloud data hosted nationally through a clickable map. It runs TerraVide under the hood and enriches your datasets. I have processed NYC's 2017 LiDAR scan (30 billion points) and host the dataset of individual Trees Cluster with digital twin models and shade metrics for the year. Its open source and downloadable for any location in NYC!
Soterria
Network of sensors which collects anonymized data on street activity and pedestrian traffic aimed to provide enterprise grade analytics to real estate and commercial businesses to improve business operations and increase revenue.
We are in the R&D phase and currently running Pilots in New York City
TreeFolio
Create digital simulation of each street tree in New York City using LiDAR and assess its local shading impact, providing data to study the distribution of street tree benefits and inform more equitable and effective tree planting strategies to reduce urban heat islands
EEG Driven Exoskeleton for Neurorehabilitation
International Conference on Robotics and Automation Engineering (ICRAE)
Singapore 2020
Acquire and interpret realtime EEG data to control a 3D printed exoskeleton to assist with movement for the physically challenged using deep learning models to classify 4 classes of intent
SYREN
Synchronous Responsive Emergency Network
Provide “green way” to incoming ambulances while simultaneously regulating traffic control by seamlessly integrating smart nodes into traffic signals across the city to create a dynamic mesh network to route vehicles on the fly.
We successfully filed 2 patents from this venture
Bus Observatory
A public archive which hosts vehicle data collected from multiple transit services in America with the aim to provide a single source for developers, researchers and hobbyist to reference, research, analyze and compare public bus services across cities. (Spring 2022)
Modeling Citizen Perception of Climate Change using Twitter Data
Tackle misinformation/fake tweets and infer public opinion on climate through sentimental analysis of twitter data and classification of filtered tweets into four types - "Factual Change, Belief of Change, Neutral, and Non-Believer"
Assess influence of demographic and neighborhood characteristics of gentrification across a city, specifically "Age, Income, Education, Cost of Living and Race" and capture changes experiences by communities through an interactive story map
Visualizing CityBike Data in NYC
Focused our analysis on ridership patterns, relationship between safety and bike lanes, classifying different types of rides, and offering informative visualizations of citi bike ridership across the city over the past decade
RIOC Bus Tracker
A local mesh network using Raspberry Pi and RF transceivers tracks Roosevelt Island bus services through a YOLO object detection algorithm with a camera module.