Prathyusha Murala

Prathyusha Murala

AI & Data Science Strategist

Scroll to explore
Prathyusha Murala

Prathyusha Murala

AI & Data Science Strategist

Bridging the gap between cutting-edge AI methodologies and real-world business outcomes, I specialize in building intelligent, data-driven solutions that unlock strategic value. My expertise encompasses the entire data science lifecycle, from advanced predictive modeling, forecasting, and personalized recommendations to natural language processing and deep learning, ensuring each solution is explainable, ethically grounded, and closely aligned with organizational objectives.

Where I’ve Worked

My journey in Data Science and AI

Mar 2022 - May 2023

Associate - Data Scientist

PricewaterhouseCoopers

NLP BERT Python PyTorch
Jun 2021 - Aug 2021

Intern - Data Analyst

Wipro

Python Django Data Visualization

Educational Background

Syracuse University

Master of Science - Applied Data Science

Relevant Coursework: Quantitative Reasoning for Data Science, Responsible AI, Information Visualization, Natural Language Processing

Dayananda Sagar University

Bachelor of Technology - Computer Science and Engineering

Relevant Coursework: Machine Learning, Deep Learning, Data Warehousing & Data Mining, DBMS

Projects

Click a chest to discover my work!

What GPT-2
Actually Does

"Hello World" but for LLMs: What GPT-2 Actually Does

NLP Machine Learning Tutorial

A plain-English explanation of how language models work, no PhD required. Covers tokenization, embeddings, attention, and the complete pipeline.

Building GPT-2
from Scratch

Building GPT-2 (124M) from Scratch in PyTorch

NLP Deep Learning PyTorch

The math, the code, and the "aha" moments from implementing a transformer. ~500 lines of PyTorch, trained on Shakespeare.

EEG Emotion
Recognition

EEG-Based Emotion Recognition with LSTM + DWT

Machine Learning Deep Learning Publication

Deep learning pipeline achieving 96.76% accuracy classifying emotions from EEG signals, outperforming SVM, KNN, and RF baselines. Published at ERCICA Conference, Springer 2023.

LexiSense
Recommender

LexiSense: Semantic Book Recommender with SBERT

NLP Recommendations Data Science

Hybrid recommender using SBERT embeddings, UMAP + K-Means clustering, and SVD collaborative filtering. 20% boost in recommendation accuracy.

World
Happiness

World Happiness Project

Data Visualization Tableau Analysis

Explored multi-year World Happiness Report & OECD data across 50+ countries. Linked social investments to a 15% rise in happiness.

UCI Adult
Analysis

UCI Adult Dataset Analysis

Machine Learning Fairness Interpretability

Predicted income brackets with emphasis on model fairness. Used LIME and SHAP for interpretability on 45K+ records.

Harmony Hub
Streaming

Harmony Hub Music Streaming Service

Data Science Database Design SQL

Scalable database and analytics solution for music streaming. Enabled real-time insights into listening habits and genre trends.

Articles

Messages from the deep - click a bottle to read!

XAI
Techniques

XAI Techniques Explained: From LIME to SHAP and Beyond

Explainable AI LIME SHAP

Explores key Explainable AI techniques, helping us understand how black-box models make decisions.

Explainable
AI 101

Explainable AI 101: Turning Black Boxes into Glass Boxes

AI Machine Learning Tutorial

A beginner-friendly introduction to Explainable AI, making machine learning models transparent and trustworthy.

Time Series
Analysis

Time Series Analysis 101: Understanding the Fundamentals

Time Series Forecasting Analysis

Covers the basics of time series data, analysis techniques, and applications for forecasting.

Gemini
MCP

How I Wired Google Gemini CLI to Control Chrome via MCP

AI MCP Automation

A hands-on guide to connecting Google's Gemini CLI with Chrome browser automation using MCP (Model Context Protocol).

Skills & Expertise

My technical toolbox built from industry experience and academic excellence

Proficiency:
Expert
Advanced
Intermediate
Basic
Beginner

Contact

Let's connect! Whether you're interested in discussing potential opportunities, have a collaboration in mind, or just want to talk data science, feel free to reach out. You can connect with me on LinkedIn or Email. I look forward to hearing from you!