EDUCATION
Indira Gandhi Delhi Technical University for Women
2023 – 2027 | CGPA - 9.14

Bachelor of Technology - Computer Science Engineering with specialization in Artificial Intelligence

TECHNICAL EXPERIENCE
1. Technology Analyst Intern | Morgan Stanley
Jun 2026 – Jul 2026 | Mumbai, India
  • •Built an AI-powered workflow tracking platform for 1,000+ batch jobs, visualizing job status, SLA, and dependencies through an interactive workflow graph.
  • •Engineered dependency graph visualization to map complex upstream/downstream job relationships, with node-level status, SLA, execution timings, and contextual job information.
  • •Implemented 5-run historical execution analysis to classify jobs as delayed, on-time, or ahead of schedule against their average start time.
  • •Developed a configurable multi-asset architecture, enabling new assets to be onboarded with minimal code changes through reusable workflow components.
  • •Built centralized failure alerting and one-click investigation, integrating AI query enhancement, relevant log retrieval, and structured analysis to streamline failure diagnosis.
    • 2. Open Source Contributor | IEEE Chapter, IGDTUW
    Nov 2025 – Dec 2025
  • •Applied Early Stopping to optimize CNN training cycles, reducing unnecessary epochs and improving overall validation performance.
  • •Performed KNN Grid Search to tune optimal hyperparameters and enhance overall prediction accuracy and model reliability.
  • •Added the missing normalization pipeline for MRI brain tumor dataset, improving training consistency and model accuracy.
  • •Cleaned and normalized intents.json file to remove duplicates, validate fields, and improve dataset integrity.
  • •Built dataset analytics to display total intents, patterns per tag, missing fields, and duplicate summary using Pandas.
  • •Performed critical fix by moving model training outside Streamlit to avoid UI freeze and reduce computation load.
  • •Added WordNet Lemmatization to improve text preprocessing and implemented a baseline Logistic Regression model to benchmark performance.
  • •Integrated Sentiment Intensity Analyzer and Toxic Input Detection for safer and more responsible chatbot behavior.

    • Evaluating Intrusion Detection using Machine Learning

  • •Evaluated traditional, neural network, and gradient boosting classifiers on CIC-CSEIDS2018 dataset to identify top-performing models.
  • •Systematically benchmarked diverse ML paradigms, from Random Forest and Neural Networks to classifiers like XGBoost and CatBoost.
  • •Achieved 93.87% accuracy on the imbalanced CIC-CSEIDS2018 dataset using a tech stack of Python, XGBoost, CatBoost, and Scikit-learn.
  • •Awarded a Certificate of Outstanding Performance for research proving the superiority of gradient boosting for intrusion detection.
  • •Presented the core research findings at the International Conference on Security, Surveillance and Artificial Intelligence (ICSSAI-2025).
    • PROJECTS
  • •Implemented a Spring Boot REST API backend that communicates with Google Gemini API, ensuring key protection via server-side handling.
  • •Tested and validated APIs using Postman to ensure reliable and consistent communication between frontend and backend services.
  • •Tools & Technologies: Java, Spring Boot, Google Gemini API, Postman, IntelliJ, VS Code
  • •Developed an NLP text classification model to identify anxiety and depression from 6,500+ social media posts using Multinomial Naive Bayes.
  • •Achieved 89.59% accuracy using TF-IDF vectorization and optimized preprocessing in a Python & Scikit-learn-based AI model.
  • •Tools & Technologies: Python, Scikit-Learn, TF-IDF, Multinomial Naive Bayes, Google Colab, HuggingFace.
  • •Led the analysis of 2000+ row sales data using SQL queries (GROUP BY, ORDER BY) to track revenue trends and identify top customers.
  • •Conducted in-depth data exploration using MySQL, generating insights through category-wise evaluations that identified key sales trends.
  • •Tools & Technologies: MySQL, Excel.
  • •Led the development of a high-accuracy (92.39%) Logistic Regression model for credit card fraud detection, enhancing transaction security.
  • •Enhanced model performance through detailed data analysis and feature selection, supporting a comparative study with Random Forest.
  • •Tools & Technologies: Python, Scikit-Learn, Pandas, NumPy, Matplotlib, Google Colab.
    • TECHNICAL SKILLS
    • •Programming Languages: Java, Python, SQL
    • •Machine Learning: Supervised classification (Logistic Regression, SVM, Random Forest), model evaluation (accuracy, precision, recall, F1-score), data preprocessing, feature engineering, NumPy, Pandas, Matplotlib, Seaborn
    • •Deep Learning (NLP): Neural Networks, LSTM / BiLSTM, text classification using TF-IDF and word embeddings, regularization techniques
    • •Natural Language Processing: Text preprocessing, tokenization, stop-word removal, vectorization (TF-IDF), fake news and sentiment classification
    • •Backend & APIs: FastAPI, REST API Development, API Testing (Postman, Linux)
    • •Databases & Data Analysis: MySQL, SQL (Joins, Aggregations), Pandas, NumPy
    • •Relevant Coursework: Data Structures & Algorithms, DBMS, OOPS, SQL, Operating Systems, Computer Networks, Artificial Intelligence, ML, DL
    CERTIFICATIONS
    ACHIEVEMENTS