Help & Supportcareers@gravityer.com
Data Science with 1+ years in Machine Learning & Predictive Analytics
AI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
M.Sc. Applied Data Science graduate with hands-on experience in machine learning, predictive analytics, and data engineering. Proficient in Python, SQL, and ML frameworks (scikit-learn, TensorFlow, XGBoost). Demonstrated ability to build end-to-end ML pipelines, perform feature engineering, and deploy interactive applications using Streamlit. Seeking a Data Scientist or Machine Learning Engineer role to apply analytical and modeling skills to real-world business problems.
SRM Institute of Science and Technology
M.Sc. · Applied Data Science
August 1, 2023 – June 30, 2025
PSG College of Arts & Science
B.Sc. · Mathematics
August 1, 2020 – June 30, 2023
Boston Institute of Analytics
Data Science Intern
December 1, 2025 – January 1, 2026
Chennai, Tamil Nadu, India
Spiro Prime Tech Services
Data Science Intern
December 1, 2024 – March 1, 2025
Chennai, Tamil Nadu, India
NLC India Limited (Navratna – Govt. of India)
Data Science Training Program
May 1, 2024 – June 1, 2024
Nellikkuppam, Tamil Nadu, India
MNIST Image Classification Using CNN
June 1, 2025 – June 1, 2026
Built and trained a CNN that classified 70,000 Fashion MNIST images into 10 categories with 93.1% test accuracy, demonstrating reliable automated image recognition suitable for retail catalog tagging or visual search applications. Reduced model overfitting and boosted generalization by ~4% through batch normalization, dropout, and data augmentation, resulting in a more robust model that performs consistently on unseen data.
E-Commerce Product Delivery Prediction
June 1, 2025 – June 1, 2026
Delivered a predictive system that flags late-delivery risk on 10,000+ orders, enabling proactive logistics decisions; benchmarked 4 models and selected Random Forest as the top performer at 0.69 accuracy. Converted raw customer, product, and logistics data into actionable features through EDA and feature engineering, then deployed the model as a live Streamlit app, giving non-technical stakeholders an interactive tool to get real-time delivery predictions without writing code.
Loan Approval Prediction System
June 1, 2025 – June 1, 2026
Built an end-to-end ML classification pipeline to predict loan approval decisions from applicant features (income, credit score, loan amount, employment history, and credit points), enabling automated, data-driven loan screening with 96% accuracy. ण्याचे ബെഞ്ച്മാർക്ക് 5 classification algorithms (Logistic Regression, Random Forest, AdaBoost, XGBoost, SVM) with feature scaling and cross-validation, then optimized the top-performing model via GridSearchCV hyperparameter tuning to maximize predictive accuracy. Serialized the final tuned model using Pickle and deployed it as an interactive Streamlit web application, allowing users to input applicant details and receive instant loan approval predictions.
Data Science
Boston Institute of Analytics
January 1, 2026 – Present
Artificial Intelligence
Boston Institute of Analytics
January 1, 2026 – Present
IBM Data Science Professional Certificate
IBM
May 1, 2025 – Present
Complete SQL Bootcamp
Udemy
April 1, 2025 – Present
Machine Learning
AWS Academy
April 1, 2024 – Present
Cultural Fit Analysis
The candidate's academic projects and internships demonstrate a strong interest and commitment to the Data Science field. The diversity of projects (e-commerce, image classification, loan prediction, sensor data analysis) shows adaptability and a broad application of data science skills. The pursuit of multiple certifications further highlights a proactive and growth-oriented mindset, which aligns well with a culture of continuous learning and innovation. The academic background in Mathematics provides a strong analytical foundation.
Soft Skills & Operational Fit
The candidate's project descriptions indicate an ability to translate technical work into business impact (e.g., 'enabling proactive logistics decisions', 'reducing false-alarm rate by ~18%'). The deployment of Streamlit apps suggests an understanding of making models accessible to non-technical stakeholders. The documentation of model performance metrics and presentation of findings to senior engineers indicate good communication and operational awareness.