Reiyo . Email & Phone Number

Jersey City, Nj, Us

New York, New York, United States

-Enhancing 3D space animations using Gaussian splats for improved swing analysis and visualisation-Developing and researching various machine learning algorithms for different use cases -Topic modelling on knowledge graphs for training LLM-Programmed a pipeline to load data from databases, perform topic modelling, and generate knowledge graphs for LLM parsing-Small Language Model: Developed a custom SLM with a Hierarchical Attention Network on a Transformer, achieving 65% accuracy on a custom dataset-2D to 3D Pose Translation: Developing an algorithm for solving the PnP(perspective-n-point) pose computation from 3D world coordinates to 2D camera coordinates to fix the data ingestion pipeline.

New York, United States

Exploratory deep learning "Proof of Concept" project was completed with senior machinelearning/data science researchers from S&P Global's Market Intelligence unit.Objective: Use of temporal and spatial climatological data from Terra and GPM satellites for purposes of training image super-resolution deep learning models that can produce high-resolution versions of General Circulation Model (GCM) climate forecast per Shared Social Pathway (SSP) CMIP6 scenarios through the end of the 21st century.Methodology: Acquired geospatial climate data from both Terra andGPM satellites. Additional GCM data was collected from the NASA Earth Exchange Global DailyDownscaled Projections (NEX-GDDP). Exploratory data analysis was performed to assess datacompleteness as well as acceptable quantitative alignment of the satellite and NEX data. Variousaggregation and rescaling methods were then applied to the satellite data for the purposes of making it suitable for use in the training of deep learning models. A collection of geography-specific super-resolution deep learning models were then developed and applied to NEX GDDP projections to produce high resolution climate forecasts per different SSP scenarios for the years 2050 and 2099.Result: A fully-developed machine learning data pipeline that supports the training, validation, testing, and deployment of geography-specific super-resolution deep learning models for geospatial climate data.

Gurugram, Haryana, India

-Conducting data-driven activities, including algorithm development, data collection, data analysis, anddashboard creation, to optimize customer interactions and inform business decisions-Unraveling insights , potential business opportunities from big data.

Masters In Ai, Artificial Intelligence

Advance Certification In Data Science, Data Science

Internship, Iot

Activities and Societies: Project Development IOT based project Development

Bachelor Of Technology - Btech, Information Technology

High School Diploma