Help & Supportcareers@gravityer.com
Data Science with less than a year in ML model development and data analysis.
AI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
Sandeep Mishra is an aspiring Data Science professional currently pursuing a Bachelor of Technology in Artificial Intelligence and Data Science. With a strong foundation in Machine Learning, Deep Learning, and statistical analysis, Sandeep has gained practical experience as a Data Science Intern at Zaalima Development, building and evaluating ML models and developing data preprocessing pipelines. Their project work showcases expertise in predictive modeling, feature engineering, and mitigating class imbalance for fraud detection.
Samrat Ashok Technological Institute
Bachelor of Technology · Artificial Intelligence and Data Science
August 1, 2022 – May 1, 2026
Zaalima Development
Data Science Intern
February 1, 2024 – July 1, 2024
India
Diamond Price Prediction
November 1, 2023 – Present
• Engineered an end-to-end regression pipeline to predict diamond prices using multi-dimensional features (carat, cut, clarity, color), achieving high predictive accuracy measured by RMSE and R2 metrics. • Implemented feature engineering and preprocessing techniques including missing value imputation, categorical encoding, and feature scaling to optimize model inputs and reduce noise. • Benchmarked multiple regression algorithms (Linear Regression, Random Forest, Gradient Boosting); selected best-performing model via systematic hyperparameter tuning with k-fold cross-validation. • Reduced prediction error significantly by iterating on model architecture and feature selection, demonstrating strong understanding of the bias-variance tradeoff.
Credit Card Risk Prediction System
September 1, 2023 – Present
• Designed a binary classification system to detect fraudulent and high-risk credit card transactions from large historical transaction datasets with significant class imbalance. • Mitigated class imbalance using SMOTE and stratified sampling strategies, improving model sensitivity (recall) for the minority fraud class by a measurable margin. • Applied statistical analysis and feature selection techniques to identify the most predictive variables, boosting model precision and recall while reducing feature dimensionality. • Evaluated model robustness using confusion matrix, ROC-AUC score, precision-recall curves, and F1-score; ensured production-readiness of the fraud detection pipeline.
Data Science Internship
Zaalima Development
January 1, 2025 – Present
Cultural Fit Analysis
The candidate's academic projects and internship align well with a Data Science role, demonstrating a clear interest and foundational experience in the field. The projects cover diverse applications (fraud detection, price prediction) and showcase a breadth of technical skills relevant to data science. The internship experience, though short, indicates exposure to real-world data science workflows. However, the candidate is still early in their career, and the overall breadth of experience is limited to academic and a single internship, which might require more mentorship in a fast-paced industry environment.
Soft Skills & Operational Fit
The candidate's project descriptions and internship experience indicate an ability to collaborate cross-functionally and ensure reproducibility and maintainability of code via Git. The academic projects demonstrate problem-solving skills and an iterative approach to model development. However, without specific behavioral assessment data, a deeper evaluation of soft skills like stress handling, leadership, or advanced communication in a team setting is not possible.