Professional Experience

I currently have four years of experience: 2\3 years of professional work experience, and 1 year of academic research experience.

I haven't gone into too much detail here about my work experience because my resume has everything. Just as a note, I can provide my resume if you don't access to it to you which has all the details if you email me at alakarthika01@gmail.com.

Current Designation - Data Scientist, Nasdaq:

I maintained and enhanced statistical models used for optimizing financial strategies, resolved real-time critical issues to ensure seamless operations. I also independently developed tools for anti-financial crime initiatives, and I have applied advanced machine learning techniques like gradient-boosting models and variational autoencoders to improve detection precision and model performance. Additionally, I worked on generative AI and LLM projects, and also processed large databases and datasets with SQL.

Academic Research with Dr. David Elbert and the Hopkins Extreme Materials Institute:

I collaborated with Dr. Maggie Eminizer to develop code for OpenMSI applications to improve data analysis workflows for material science research. I also worked on deep learning and clustering techniques to analyze material science datasets, achieving really high accuracy in identifying patterns. I also designed and implemented an image processing framework with least-squares optimization to automate flyer curvature analysis, and streamlined data processing for over 10,000 experiments.

Educational Background

Master of Science in Engineering, Data Science - The Johns Hopkins University

Academic Projects:

Machine Learning-Assisted Acceleration of Crystal Synthesis

This was my thesis. As part of my thesis at the NSF PARADIM User Facility, I applied machine learning to optimize crystal synthesis for quantum materials using advanced optical floating zone furnace (FZF) methods. I developed a clustering-based approach to navigate FZF’s complex parameter space, addressing challenges like high-latency responses and balancing multiple control parameters. Additionally, I automated the computation of flyer curvature and tilt for Flyer-Velocity Analysis experiments in the Laser Shock Lab, contributing to the creation of spall-resistant aluminum. This work is to enhance precision in material synthesis and advanced experimental efficiency in quantum material and aerospace research.

Twitter Analysis of Attitudes towards Gender with NLP

My team and I conducted an in-depth analysis of 2022 Twitter data, before it was acquired, to understand online attitudes towards gender. Using Natural Language Processing (NLP) techniques, we analyzed and clustered tweets to extract insights into gender perceptions. The project involved sentiment analysis and tweet generation with data collected via the Twitter API, revealing an overall negative bias in sentiment towards female-oriented tweets. This analysis highlighted how language on social platforms reflects societal attitudes and biases where further research in online behavior and gender studies could be done.

Detection of Atrial Flutter and Ventricular Tachycardia using ML

I worked on applying machine learning techniques to differentiate atrial flutter (AF) and ventricular tachycardia (VT) using ECG signals. This project utilized dimensionality reduction techniques and unsupervised ML algorithms, alongside Wavelet Transformations, to study patterns in these cardiac conditions. I implemented deep learning models, including transformers, CNNs, and RNNs, and optimized their performance through grid search with supervised techniques.

Bachelors in Engineering, Electronics and Communication Engineering - College of Engineering, Anna University

Academic Projects:

Design of IoT System to Measure Soil Parameters with Crop Prediction ML

I designed an IoT-based system to monitor and detect soil parameters, using Arduino IDE and a NodeMCU. The system collected real-time data to determine whether an area was being adequately irrigated. Using Python, I implemented a machine learning module with KNN, SVM, and Decision Tree algorithms, achieving 91% accuracy in predicting the most suitable crop to plant based on the detected soil conditions. This project combined IoT and machine learning to provide a practical solution for agricultural management.

Hypothesis Tests Using Shiny

I developed a Shiny App to perform statistical hypothesis testing on various R datasets, including Iris, Plant Growth, and Tooth Growth. It allows users to conduct hypothesis tests on pre-loaded datasets, showcasing statistical results and visualizations.