Divyanshu Singh
AI/ML intern with hands-on experience in Python programming, data analysis, and machine learning algorithm implementation across supervised learning, deep learning, and computer vision.
Skilled in data visualization and model assessment methodologies (accuracy, precision, recall, F1-score, ROC-AUC, confusion matrix), with a track record of building end-to-end ML pipelines from data preprocessing through deployment-ready evaluation.
Experience
AI/ML Intern
3Skill
Remote
Developed an AI-Based Hiring Prediction System using Python programming and Scikit-learn to automate candidate selection through predictive analytics and machine learning algorithms. Engineered robust data preprocessing, feature engineering, and model optimization pipelines as part of end-to-end data analysis workflows to enhance classification performance.
Built and evaluated machine learning models (Logistic Regression, Random Forest, XGBoost) using model assessment methodologies including Accuracy, Precision, Recall, and F1-Score. Used data visualization techniques (Matplotlib, Seaborn) to communicate model performance and dataset insights alongside Pandas, NumPy, Scikit-learn, and Git/Git Hub.
AI/ML Intern
AICTE IBM SkillBuild
Remote
Developed a Fake Job Detection System using Natural Language Processing (NLP) and machine learning algorithms to identify fraudulent job postings. Applied advanced text preprocessing, TF-IDF vectorization, feature extraction, and exploratory data analysis (EDA) on real-world datasets.
Trained and optimized classification models using Scikit-learn, applying rigorous model assessment methodologies for reliable fraud detection. Strengthened expertise in Machine Learning, Deep Learning fundamentals, NLP, Data Analytics, Model Validation, and AI-driven Problem Solving through hands-on Python programming.
Education
Onkarmal Somani College, Jodhpur
Bachelor of Computer Applications
Skills
PythonJavaJavaScriptCC++RHTMLCSSMachine Learning AlgorithmsCNNsRNNsData PreprocessingNLPNatural Language UnderstandingLLMsRAGTransformersExploratory Data AnalysisData VisualizationNumPyPandas
Projects
AI-Based Hiring Prediction System
Developed an AI-powered Hiring Prediction System using Python programming, Pandas, NumPy, Scikit-learn, and machine learning algorithms to classify candidates as Hired or Rejected based on resume attributes. Performed data preprocessing, feature engineering, label encoding, exploratory data analysis (EDA), and feature selection to optimize model performance and data quality.
Trained and evaluated Logistic Regression, Decision Tree, and Random Forest models using model assessment methodologies (Accuracy, Precision, Recall, F1-Score, Confusion Matrix, Cross-Validation) to build an automated resume screening solution.
Fake Job Detection System
Developed a Fake Job Detection System using Python, Scikit-learn, NLP, and machine learning algorithms to classify job postings as Fraudulent or Legitimate from structured and textual data. Applied text preprocessing, tokenization, TF-IDF vectorization, feature engineering, and data cleaning to transform job descriptions into meaningful numerical features.
Implemented and compared Logistic Regression and Naive Bayes models, optimizing performance through hyperparameter tuning and model assessment methodologies (Accuracy, Precision, Recall, F1-Score, ROC-AUC, Confusion Matrix).
Customer Churn Prediction System
Developed a Customer Churn Prediction System using Python, Pandas, NumPy, Scikit-learn, and machine learning algorithms (Logistic Regression, Decision Tree, Random Forest) to predict customer attrition with high accuracy.
Performed data preprocessing, EDA, feature engineering, label encoding, feature scaling, and model assessment (Accuracy, Precision, Recall, F1-Score, ROC-AUC, Confusion Matrix) to optimize prediction performance.
Credit Card Fraud Detection System
Developed a Credit Card Fraud Detection System using Python, Pandas, NumPy, Scikit-learn, machine learning algorithms, XGBoost, Random Forest, and Logistic Regression to classify fraudulent and legitimate transactions in highly imbalanced financial datasets.
Applied data preprocessing, feature engineering, SMOTE (class imbalance handling), feature scaling, hyperparameter tuning, cross-validation, ROC-AUC, precision-recall analysis, confusion matrix, and model assessment methodologies to maximize fraud detection performance and minimize false positives.