Data Scientist

👨‍💻 Work Globally

Key Expertise

• Evaluate advanced statistical models and approaches for application
• Feature engineering and devising parameters
• Constructing knowledge base for machine learning models
• Building deep neural networks and their hyper-parameter tuning
• Deep development understanding

Tools

Python, PySpark, Keras, Tensorflow, PyTorch, Git, Jupyter, OpenCV, Numpy, Scikit, Pandas, Matplotlib, Spyder, DataBricks

Summary

I am a Data Scientist who is intrigued by the idea how data shapes our lives and curious to all fields where data has big impact. I like to work on projects which allow me to combine multiple approaches to get deeper insights and hidden answers. I have rich experience in areas working in Computer Vision, Time-series modelling, Audio engineering, Language processing and more.

I firmly believe that Feature Engineering and building derived parameters are paramount for getting the most out of raw data and are amongst the most important features for building efficient deep learning networks. I have built knowledge base for some of my projects which helps the program to remain updated to the emerging market trends and analyzing new ones. Exploring and understanding the vast amount of Neural networks that exist in today's world is an absolute must. Each have their own merits and demerits based on the type and structure of data as well as the goal they are built to achieve. I have worked with LSTMs, CNNs, GANs, xGBoost and some older machine learning models like SVM, Logistic/Linear Regression, K-means clustering and more. For attaining the maximum output from these networks, one has to find the best set of hyperparameters to be used to train the model and I have used K-Fold Cross Validation, GridSearch for tuning.

Work History

Data Scientist

FanClash – (Aug 2022 - Present)

• Constructed ML Fraud Detection models using conventional algorithms of Geometric models like IForest using Tensorflow ****as well as GraphDB ****solutions using AWS Neptune.