UMESH CHANDRA
Data Scientist & Machine Learning Specialist
I am a **passionate and results-driven Data Scientist** with a solid foundation in statistical modeling, machine learning, and data visualization. I leverage Python and cloud technologies to transform complex datasets into actionable business insights and drive measurable impact.
Education
B.Tech (Electronics and Communication)
Engineering Degree
[Institution Name Placeholder]
[Year Placeholder]
The Electronics and Communication background provides a strong foundation in **signal processing, mathematics, and complex systems**, which are highly valuable skills for advanced machine learning and data science applications.
Technical Expertise
Programming
Python (Pandas, NumPy, Scikit-learn), R, SQL, Bash.
ML & Stats
Regression, Classification, Clustering, Time Series, A/B Testing, NLP, Deep Learning (PyTorch/TensorFlow).
Data & Cloud
PostgreSQL, MongoDB, ETL, AWS (S3, EC2), Docker, Git/GitHub.
Visualization
Matplotlib, Seaborn, Plotly, Tableau, Dashboarding.
Featured Data Science Projects
1. Predictive Customer Churn Model
Developed a highly accurate classification model (XGBoost) to predict customer churn for a telecom company. Engineered features from transactional and demographic data, achieving an AUC score of 0.89. The model provides insights into key factors driving churn, enabling targeted retention strategies.
Tools: Python, Pandas, Scikit-learn, XGBoost, Matplotlib, AWS S3.
Key Takeaway
89%
AUC Score for Churn Prediction
2. Social Media Sentiment Analysis Engine
Built an end-to-end NLP pipeline using BERT and TensorFlow to classify millions of social media comments into positive, negative, and neutral sentiments regarding a major brand. Deployed the model as a scalable API. This project included data scraping, cleaning, tokenization, and fine-tuning the deep learning model.
Tools: Python, TensorFlow, BERT, Hugging Face, Flask, Docker.
Model Used
BERT
Deep Learning Architecture
3. Interactive Sales Performance Dashboard
Designed and deployed an interactive dashboard using Plotly Dash to monitor real-time sales performance metrics. Implemented robust ETL pipelines to clean and aggregate data from multiple SQL sources daily, providing key stakeholders with a single source of truth for decision-making.
Tools: Python, Dash/Plotly, PostgreSQL, Pandas, Docker.
Pipeline
ETL
Daily Automated Data Processing
Get In Touch
I am actively seeking new opportunities and challenges in the field of Data Science and Machine Learning. Feel free to reach out to discuss potential collaborations or job roles.
Email Me: chandraumesh796@gmail.com