Atharv Tekurkar

About Me 👨

I am an international student from India at Union College pursuing a Bachelor of Science degree in Mathematics and Computer Science, maintaining a GPA of 3.7 and achieving the Dean's List for three consecutive years.

My professional journey includes a Data Analyst internship at Bajaj Allianz Life Insurance Co. Ltd., where I worked towards the development of a machine learning-based recommendation system to provide personalized life insurance suggestions, significantly enhancing the customer experience.

Additionally, during my time as a Quantitative Analyst Intern at CRISIL Ltd., I gained hands-on experience in developing, validating, and testing various quantitative models, achieving over 97% accuracy. My role involved rigorous data analysis, model validation, and the creation of data visualizations to effectively communicate key insights.

I am passionate about using my computational and analytical skills to solve complex challenges and continuously grow my knowledge in the applications of mathematics and computer science.

Apart from the academics, some other things I enjoy are:

Cricket
Basketball
Chess
Weightlifting
Trekking

Projects 💻

As an aspiring data analyst, most of my projects inlcude collection and analysis of data and building ML models for forecasting or predictions. Below are a few highlights from my portfolio:

Predicting bank failure using ML - Altman Z-Score & Stock Price Volatility

Built predictive models using 24 years of data for 26 public U.S. banks (13 failed & 13 non-failed), incorporating stock price volatility, Altman Z-Score trends, and other micro and macroeconomic indicators.
Achieved 96% predictive accuracy using XGBoost, outperforming Random Forest and Neural Networks, to build a model to forecast bank failures in advance, enhancing risk assessment and early action.

Predicting cryptocurrency prices using ARIMA

Analyzed 900k+ data points of the cryptocurrency market and built a Python statsmodel ARIMA model for short-term price prediction of Bitcoin with an accuracy of 98%.
Evaluated and matched price trends for the top 100 cryptocurrencies to build a weighted portfolio, and developed a model to predict its short-term price movements.

Predicting weather and climate change using RNN: The PDE Approach

Developed a climate forecasting model using Recurrent Neural Networks and partial differential equations, achieving 91% accuracy in predicting key climate variables, such as temperature, precipitation, etc.
Optimized the machine learning algorithms by training the model with 9-month data (1991 - 2020) over 1000 epochs with TensorFlow, and accurately predicting climate conditions for 3 months.

Photography 📷

I am very passionate about photography! Most of the photos I have are taken in my mobile phone, but have a very professional look. Below are some of my personal favourites from around the world.

Rajasthan, India

Mt. Machhapuchhare, Nepal

Sydney, Australia

Cinematography 📹

I also have a youtube channel where I post cinematic creations, vlogs, and other informative videos when I get a chance. Here's a sneak peak of one of my cinematic creations that I am proud of.

Connect 📧

Thank you for visiting my page! If you have any feedback or wish to leave me a message please free to use the form below to send me an email and I will get back to you!