About
I am a Data Scientist at the Johnson Controls, .
I graduated with a Masters in Business Analytics from UIUC, with a strong background in Data Science, Data Modelling and Machine Learning. Coming from a traditional Data Analytics background, it has enabled me to excel in the field of Data Science and Machine Learning by advancing skills, always innovating to contribute to the open-source community.
Detail-oriented and high-energy Data Scientist with over 5 years of experience in advanced analytics and machine learning. Expertise in building and deploying machine learning models, and working with large datasets using data science tools. Proven ability to create derive actionable insights through statistical analysis and data visualisation. Strong communicator skilled in conveying complex technical information to diverse stakeholders, driving key business decisions, and fostering collaboration. Committed to continuous learning and delivering high-quality results.

Data Maverick
- Phone: +1(447)902-0478
- Location: United States
- Email: pratik.relekar@gmail.com
Skills
I possess a diverse skill set grounded in data science, data engineering and analytics, harnessing the power of algorithms, statistics, and predictive modeling as well as algorithmic data modelling to uncover insights and drive business growth. Following Skill section showcases proficiency with respect to the project objectives achieved, referencing variety of web sources and literature:
Resume
I am passionate about creating scalable, efficient, and user-friendly visualization dashboards and data science solutions that solve real-world problems and add value to the society.
Education
Masters, Business Analytics
2022 - 2023
University of Illinois at Urbana-Champaign, Champaign, IL
Relevant Courses: Enterprise Database Management, Big Data Analytics, Business Intelligence, Enterprise Database Management, Big Data Infrastructures, Financial Database Management and Analysis, Supply Chain Analytics
Bachelor of Engineering, Electrical Engineering
2010 - 2014
University of Pune, Pune, India
University Position
Research Assistant
September 2022 - May 2023
Data Science Research Services, UIUC, IL
- Employed Optical Character Recognition (OCR) to extract text data from 2000+ financial statements and credit reports of 40+ companies. Also performed sentiment analysis using Masked Language Models (MLM) on the text extracted from financial statements.
Professional Experience
Data Science, Global Products and Innovation
January 2025 - Present
Johnson Controls, Milwaukee
Data Scientist (Research)
June 2023 - January 2025
Data Science Research Services, USA
- Invented open-source parallel computing Python library for NielsenIQ Retail Scanner Data, reducing data reading time by 92%, accelerating overall processing speed by 3x, and cutting server costs by 20%
- Improved investment strategies and risk management by analyzing financial statement sentiments using BERT and other NLP models, achieving 83% accuracy and providing actionable market insights.
- Spearheaded web scraping of 2000+ 10-K SEC filings and employed Optical Character Recognition to transform them into text.
- Supported strategic financial management by transforming 2,000+ tax documents and 10-Ks into machine-readable text through an engineered OCR pipeline, significantly improving data accessibility and facilitating in-depth research on tax consultants.
- Developed Docker container featuring front-end interface to facilitate easy retrieval of information from financial documents.
- Derived the relationship between political/weather events and customer buying trends using Nielsen’s data with Dask on a 576 GB cluster, aiding businesses plan inventory, marketing, and staffing needs.
Business Analyst
November 2021 - July 2022
Johnson Controls, India
- Enhanced data management efficiency by 30% and reduced retrieval times by 40% by developing and deploying an ETL pipeline to migrate data from Hadoop to SQL Server, streamlining data workflows.
- Improved customer identification accuracy by 20% and increased potential sales lead identification by 30% by implementing Named Entity Recognition using Fuzzy logic to clean 1 million CRM records, enhancing data integrity through deduplication.
- Optimized SQL queries powering Power BI dashboards for sales, reducing refresh times by 20% and boosting sales activities by $50M.
- Streamlined Agile project management using JIRA and led data analysis projects, improving process flows by 75% and clearly communicating data insights to non-technical stakeholders
Data Analyst/Business Intelligence Analyst
June 2021 - October 2021
Indium Software, India
- Increased client engagement and conversion rates by 15% by developing over 50 Tableau dashboards, delivering key customer insights and improving visibility for data-driven decision-making.
Data Analyst
November 2019 - May 2021
Skymet Weather Services Ltd. (Dept. of Agriculture, India)
- Reduced querying time by 75% by migrating agriculture data from flat files to PostgresSQL and engineering a user-friendly front-end interface, optimizing data retrieval and improving user experience.
- Implemented machine learning algorithms including Linear Regression, KNN regression, Random Forest, and Neural Networks to predict agricultural production, informing design of crop-advisory policies, forecasting defaults across 25+ districts.
- Performed comprehensive Exploratory Data Analysis (EDA) to pinpoint essential weather and soil variables, applied feature engineering methods such as one-hot encoding and normalization, markedly enhancing model accuracy.
- Engineered soil health classification model utilizing Logistic Regression, Decision Trees, Random Forest, SVM and Boosting achieving 85% accuracy to improve soil management strategies.
Design Engineer/Electrical Engineer
January 2015 - October 2019
Bharat Forge Ltd & Trimurti Stampings, India
- Leveraged electrical engineering expertise in various industries from 2015 to 2019, building automotive systems design.
Projects
My portfolio showcases a diverse range of projects spanning from applications of LLM, Data Optimization, Natural Language Processing and Machine Learning solutions. These projects highlight my proficiency in creating versatile and impactful data solutions across different domains. NielsenIQ Retail highlights my innovative ability to solve business and research problems by introducing first of its kind and state of the art Python library that enables researchers and data scientists optimize big data reading (terabyte sized) process at lightning fast speeds.
- All
Contact
I'm excited to connect and engage in conversations related to Data Science, Deep Learning, Applied Mathematics and innovative projects. Let's explore new possibilities together!
Location:
United States
Email:
pratik.relekar@gmail.com