Data Scientist | Software Engineer.
Hi, I am a Software Engineer who loves Data Science and building products related to data. I recently analyzed Stackoverflow Annual Developer Survey to answer some business questions. I also built a web app that classifies messages in appropriate categories for disaster response. I'd love to combine my passion for data science with my software development skills to continue building products for people.
First Year: University of Groningen, Netherlands
Second Year: University of Basque Country, San Sebastian
Major: Natural Language Processing Applications
Relevant Coursework: Machine Learning, Linguistic Analysis, Statistics, Corpus Linguistics
Relevant Coursework: Object Oriented Programming, Data Structures and Algorithms, Software Engineering, Software Construction, Machine Learning, Software Design and Architecture
Relevant Coursework: Introduction to Data Science, Software Engineering, Data Engineering, Experimental Design and Recommendations
Implemented various classic Convolutional Neural networks like LeNet-5, AlexNet, VGG-16 and ResNet-34 using Pytorch. Tested these models on MNIST, FashionMNIST and CIFAR-10 Datasets - Github
Worked on Unsupervised Water Body Classification using Deep Learning (Keras and Tensorflow). F1 Score of 85% on classification of water bodies in EuroSAT
Worked on a DAAD-funded project titled, ‘Forest Monitoring and Change Detection using Remote Sensing Imagery’
Built an unsupervised pipeline for forest regions binary classification and achieved an f1 score of 91% on the test set of EuroSAT.
I can provide you with production ready code for the following services
My projects in data analysis, data modeling, web scraping and web development
Using Python's libraries numpy, pandas, matplotlib & scikit-learn analyzed the dataset of past three years to answer questions like the effect of education on salary & job satisfaction.
Classified remote sensing forest regions in a progressive unsupervised way due to lack of annotated data and surpassed state-of-the-art unsupervised classification results for classification on Forest regions using satellite imagery
Build ETL and ML pipelines for the data loading, data transformation and classification of messages into disaster categories. The dataset was from FigureEight.
Built a Quora Scraper using selenium to automate the login process and scrape questions/answers against any query.
An online Python coding challenges platform like Hackerrank built using the MERN stack and Hackerrank API.
Various Web Scraping Programs in Python for static as well as dynamic websites along with automating the authorization process.
My blog posts on medium regarding my projects and learning
Implementation of various recommender systems' algorithms for recommending articles on the IBM Watson Studio Platform.
Implementation of ETL and ML pipelines for data loading and classification of messages for disaster response
A data-driven approach to gain insights from Stackoverflow Annual Developer Survey Dataset from 2017 to 2019
I am available for any freelancing work related to my services and you can contact me on my email or click below to hire me on Upwork