Hi, I'm Chandan Rudrappa.
A
Self-driven and detail-oriented data scientist with a strong analytical mindset, passionate about building scalable ML solutions and solving complex real-world problems through data.
About
Data Science graduate from The University of Texas at Arlington. I enjoy turning messy data into clear insights and building end-to-end solutions that create measurable impact. I have hands-on experience with Python, PyTorch, TensorFlow, Streamlit, Flask, SQL, and big-data tools like Spark and Hadoop. My industry and research work spans NLP (TextCNN, Transformers), predictive modeling, dashboard development, and data automation. I’m committed to building scalable, meaningful applications—whether that’s a high-accuracy ML model, an automated analytics tool, or a streamlined data workflow.
Experience
- Developed a Streamlit application that ingests waiver documents in Excel format and automatically highlights content discrepancies, reducing manual review time by 50% and improving validation accuracy.
- Built and deployed TextCNN and Transformer-based classifiers to categorize document context into 20+ tag categories significantly speeding up downstream document processing.
- Tools: Python, Streamlit, Tensorflow, Keras, Huggingface Transformers, Pandas, NumPy
- Tutored 100+ undergraduate students in Statistics, Probability, and Calculus I–III across two tutoring centers, improving student performance by an average of 20%.
- Supported service-learning data projects with community partners, providing analytical and technical assistance to teams of 10–15 students.
- Cleaned, transformed, and validated real-world community datasets, increasing data usability and enabling students to perform meaningful statistical analyses.
- Guided students on research methodology, hypothesis testing, and statistical interpretation, helping translate community-driven questions into actionable insights.
- Tools: Python, Excel, R
- Designed and developed 150+ reusable data science functions for the LEAPS analytics platform, enabling faster implementation of data visualization, data management, and time series analysis workflows.
- Upgraded 80+ static matplotlib visualizations into interactive Plotly dashboards, increasing user engagement and reducing dashboard creation time by 25%.
- Involved in building a recommendation engine to enhance the LEAPS user journey by providing personalized data science problem suggestions, improving platform retention by 15%.
- Developed an internal employee management portal using Flask and Django, automating record tracking and improving data accessibility across teams
- Authored 60+ Jupyter notebooks for applied data science case studies, helping users learn key ML concepts through hands-on experimentation.
- Collaborated with cross-functional teams to create 5 instructional data science courses, simplifying complex topics and improving learner comprehension scores by 30%.
- Led client open house sessions for Siemens and Aditya Birla Group, presenting classification and regression analyses to groups of up to 30 participants, strengthening client engagement and understanding.
- Tools: Python, Scikit-learn, Pandas, NumPy, Matplotlib, Plotly, Jupyter
Projects
A Streamlit app that automatically compares waiver documents and highlights content discrepancies.
- Built a Streamlit application that ingests Excel waiver files and highlights differences between versions.
- Reduced manual review time by 50% through automated document comparison and structured output.
- Integrated preprocessing and formatting pipelines to ensure consistent validation across diverse document types.
Predictive modeling project on tensile test outcomes using industrial manufacturing data.
- Selected from 40+ teams to work directly with Scot Forge engineers on a real-world dataset.
- Performed EDA, hypothesis testing, feature engineering, and preprocessing.
- Built a classification model achieving 98.82% accuracy after resampling.
- Identified key predictors influencing tempering effects on tensile strength.
Computer vision integration for autonomous rover navigation and object detection.
- Implemented ArUco marker tracking and LiDAR-based object detection modules.
- Integrated ZED stereo camera SDK for depth-based localization.
- Configured and optimized the NVIDIA Jetson TX2 using the latest JetPack SDK.
NLP models to classify waiver content into 20+ tag categories using deep learning.
- Developed TextCNN and Transformer-based classifiers for context tagging.
- Improved tagging efficiency and reduced manual processing time significantly.
- Enabled structured downstream automation for document review pipelines.
Statistical analysis of IMDb’s Top 250 movie ratings using a Goodness-of-Fit test.
A GUI-based application for managing and updating database records using Python and SQLite.
Skills
Programming Languages & Databases
Python
C
C++
R
MySQL
SQLite
Data Science & ML Libraries
NumPy
Pandas
Scikit-Learn
Matplotlib
Plotly
OpenCV
Machine Learning & Big Data Frameworks
TensorFlow
PyTorch
Keras
Hadoop
Spark
Streamlit
Cloud, DevOps & Collaboration Tools
Git
Docker
Kubernetes
Jenkins
AWS
Bitbucket
Education
University of Texas at Arlington
Arlington, USA
Degree: Master of Science in Data Science
CGPA: 4.0/4.0
- Neural Networks
- Computer Vision
- Big Data
- Probability & Statistics
Relevant Courseworks:
Ramaiah Institute of Technology
Bengaluru, India
Degree: Bachelor of Engineering in Electronics and Communication
- Introduction to ML/DL
- Data structures using C++
- Cryptography
- Internet of Things
Relevant Courseworks:


