Kumar Ayush

Software Engineer

Trinity Mobility Private Ltd.•  October 2021 - Present

1. ML Composer Tool – AutoML | Scikit-learn, Flask, Apache Spark, Seaborn - Developed the pipeline for implementing the AutoML functionality which collaborates multiple algorithms and fine-tuned the parameters using GridSearchCV - Model Pipe line such as Prediction, Forecasting (Univariate and Multivariate), Anomaly Detection, Recommendation, Text Analytics (Text Classification, NER, zero shot Classification) are part of the tool - Array of data pre-processing, EDA and other validation techniques for better visualization and understanding of data set, with possibility of customization 2. Digital Twin – Smart Commnity | Jupyter Notebook, Keras, Flask, FbProphet - Developed several machine learning models to predict the anomaly in the mechanical system in the building, forecast the AQI, Energy ect.. for real time monitoring of the building 3. Predictive Policing – Crime Forecasting | Jupyter Notebook, Keras, Flask, Seaborn - Developed the classification model to predict the crime areas in the nearby future based on historical data taken from Dial 100 system - Developed a classification model to predict the probability of different types of crime events that might happen.

Systems Engineer

MAN Energy Solutions•  June 2018 - October 2021

1. Digital twin technology for existing turbine operations - Collect, clean, transform and validate data, part of data pre-processing - Feature selection (RF, Stepwise first order) and Feature extraction using machine learning technique and domain knowledge - Presents data in the form of charts, plots and tables for immediate reference – Descriptive analysis - Model validation using K-fold Cross Validation, Error metrics. Identify effect(trend) of individual variables through sensitivity analysis - Develop machine learning predictive models (Linear, Random Forest, MARS, Stepwise first order, KNN, PLS) to accurately predict the behavior of turbine process parameters - Work closely with software development teams to ensure accurate integration of machine learning models into firm platforms. 2. Development of Valves and Actuator Localization - R&D for Emergency Stop Valve - Localization of hydraulic Actuator for Emergency Stop and Control Valves - Project Management for the New Project Development - Failure mode effects analysis (Design and Manufacturing) - Vendor Qualification and process audit

Design Engineer

Tata Consultancy Services Ltd•  April 2014 - May 2018

1. Design Engineering for the Turbine Components: (Mitsubishi Hitachi Power Systems ) - Layout and general arrangement drawings for the core machine - Generic models and their assembly based on the variable parameters - Material selection based on the requirements and the usage type - Part of Design Failure Mode Effect Analysis (DFMEA) team for the evaluation of the design and detailed design

Great Lakes and University of Texas at Austin

Artificial Intelligence and Machine Learning, Post Graduate Program•  May 2020 - September 2021

Post Graduate Program in Artificial Intelligence and Machine Learning - Forecasting Vaccination Drive in India for Herd Immunity using SIR and Prophet Model [ IEEE World AI IoT Congress 2021] - Ergonomic in Sports through Human Pose Estimation using Computer Vision

Sri Siddhartha Institute Of Technology

Mechanical Engineering, B.Tech•  August 2009 - September 2013

B.Tech in Mechanical Engineering