Kshitij soni

Kshitij soni

Data Science & AI Strategy Leader · Data Science Club

United States

In My Own Words

With over a 10 of experience in data analytics and a robust foundation in AI and machine learning, I specialize in empowering organizations to leverage advanced AI technologies for strategic decision-making and innovation. My academic journey began at the Indian Institute of Technology (IIT), where I earned a Bachelor of Technology in System Science, followed by a Master's in Business Analytics from the University of Texas at Dallas (JSOM).

As a Senior Analytics Consultant at EXL, my role transcends traditional data analytics, venturing into the realm of artificial intelligence. I've been instrumental in designing and deploying transformative solutions across finance and healthcare sectors, leveraging not only big data technologies and predictive modeling but also advanced AI frameworks including Large Language Models (LLMs) and Retrieval-Augmented Generation (RAG). My proficiency extends to a comprehensive toolkit comprising Python, SQL, Hadoop, Spark, alongside deep learning frameworks essential for AI-driven analytics.

My passion lies in the intersection of data science and AI, where I continuously explore and integrate the latest advancements such as transformer models and neural network architectures to enhance the intelligence and efficiency of analytical models. Collaborating with multidisciplinary teams, I thrive on driving impactful innovations that redefine industry standards and create tangible business value.

I am eager to connect with forward-thinking professionals and organizations that are at the forefront of harnessing AI to unlock new opportunities and drive digital transformation. Let's explore how we can leverage the power of AI and data analytics to achieve unprecedented outcomes in the ever-evolving business landscape.

Work Experience

UT Dallas- Volunteer

Data Science Club

2018 – Present

Data Scientist

Altisource Labs

2016 – 2018

Designed model selection and feature engineering on more than a million of rows of data, thus decreasing memory space of features matrix by 80% and increasing forecast accuracy by 10%. • Generated, tests hypotheses, analyzes and interprets the statistical results on performance evaluation of the two software. • Led a team of 2 in analysis, architecture, design, and development of US house market trend prediction project. • Developed the model with error rate of 5% using time series analysis and multivariate regression techniques.

Data Science Intern

Serendio Software

2015 – 2015

Developed the recommendation engine by pre-processing the e-commerce data to predict the buying behavior with an accuracy of 96%. • Built Text Analysis framework in Python to extract topics, categories, sentiments and keywords from unstructured text regardless of its source, using Stanford CoreNLP tools.

ACADEMIC PROJECT Database Management for Business Analytics • Imported, Exported, manipulated, performed analysis on provided university database using SQL and NOSQL.

Secretary

IIT J

2014 – 2015

Vice Captain

IIT J

2014 – 2015

Education

Bachelor of Technology

Indian Institute of Technology

2012 – 2016

Bachelor of Technology

M.S.

The University of Texas at Dallas

M.S.