Avatar

Hi! I'm Aditya Kaushal

I'm currently a Data Engineer who’s previously interned/worked at Hitachi as a Data Analyst Intern and Aerogram (IIT-Delhi Startup) as a Data Engineer Intern. My primary focus is on Data Engineering, Big Data Analytics and Cloud. Previously I finished up my B.Tech/B.E in Computer Science and Engineering in Spring 2020, at Chandigarh University. My goals are to learn as much as possible from my experiences and interactions with others, and leave a positive impact on the world.

AirSol

Time Series Forecasting and regression analysis web app to predict and forecast PM 2.5 values and other Air pollution metrics 3 hours ahead. This project uses the Facebooks Prophet algorithm to forecast the PM 2.5 values 3 Hours ahead.

Face2Gene

Facial Recognition or Facial Image Detection can be used to identify the individuals using their facial features. Facial Recognition or Face Detector can be used to know whether a persons Identity through his/her face features. This Repository can be used to carry out such a task. It uses your WebCamera and then identifies your expression in Real Time.

Technical Skills: OpenCV, Python, Scalar Vector Machine, K-Fold Cross Validation, SciPy, NumPy, Pandas, Matplotlib

Loan Prediction Analysis

Carried out Analytics on Loan Predition Dataset to predict whether the loan can be granted or not. This Repo utlises the Statistical Plots and other visualisations to understand the dataset better.

Technical Skills: Python, Seaborn ,NumPy, Pandas, Matplotlib

Range of incubation periods for Covid-19 in Humans

Processed 17+GB of Dataset consisting of 200,000+ scholarly articles provided by AI2, Georgetown, NIH & The White House to mine text related to incubation period of COVID-19. Utilised the Python script to parse JSON files to extract days and weeks of incubation and plotted a histogram for visual representation.

Technical Skills: Python (Data Analytics, visualisations and data mining)

Finding Lanes

Finding Lanes using OpenCV, for simulation of self-driving cars.

Technical Skills: OpenCV, Python, Numpy, Canny Edge Detection

Canvas Sketcher

Flask Web App for creating Pencil sketch.

Technical Skills: Python, Flask, OpenCV, Requests

PM2.5 Forecaster

Time Series Forecasting and regression analysis web app to predict and forecast PM 2.5 values and other Air pollution metrics

Technical Skills: Python, Numpy, Pandas, Matplotlib, Seaborn, Exponential Moving Averages,

Machine Learning: Cost Functions and Gradient Descent

A Machine learns through a model, and a model is basically a function f such that f(x) maps to y. The model learns through taking features classes or we can say, input variables which are also known as X. These are usually independent variables. These are used to predict the target variables that are Y (known as the response or more traditionally the dependent variables).

IoT Enabled Blockchain and Machine Learning and its Possibilities in Supply Chain Management.

IoT enabled Blockchain to enable secure tracking of certifications and other important information in supply chains in a secured and shared Hyper-ledger which is transparent, tamper-proof, and distributed across multiple computers that are running by consensus i.e., with trust and reliability.The Supply chain can be tracked by using these above-mentioned tags in two ways. It can store data directly, and will also be able to update as the items move through the Supply Chain exchanging hands.

Aerogram (IIT Delhi)

Utlised MQTT protocols for migrating telemetry feed from low bandwidth IoT-Devices to Google IoT Core for further data collection and accumulation. Developed ETL Pipelines to receive and send telemetry feed from Google Cloud Pub/Sub to store data into Google Cloud SQL and Google Firestore by leveraging Google Cloud Functions using Python. Developed a Flask Web App for forecasting PM2.5 values. Developed Python scripts for publishing and subscribing data and payload messages for testing and understanding the working of Google Pub/Sub and Google IoT core.

Technical Skills: Python, NumPy, Pandas, Matplotlib, Seaborn, Cufflinks, Flask, Google Cloud Platform (Google Pub/Sub), HTML, CSS, JavaScript, SQL, Tableau

APIs: Pygal, Facebooks Prophet for Forecasting

Hitachi India

Developed a Data Scraper using Python by utilizing Pandas, ScraPy, SciPy, NumPy and Python-based Automation(Selenium) and extracted unstructured Data from Web Pages and simultaneously stored the Data in CSV formats. Assesed the accuracy and consistency of the Arrival Time of Trains on the National Train Enquiry System.

Technical Skills: Python, WebScraping, Excel for Analytics, Numpy, ETL, Data Extraction, Data Handling