Hey, I’m Khushbu Chaudhary, M.Tech CSE @ IIT Hyderabad. I work on ML, Deep Learning, Domain Adaptation, Medical Image Segmentation, and AI-systems + exploring.
AI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
Capstone-2
June 4, 2026 – Present
LDA + Hawkes Process framework for predicting future research topics and recommending citations in scientific literature. Uses Weibull temporal kernel and citation-supervised attention over frozen LDA representations.
View Projecthuman-value-integrated-agent
April 22, 2026 – Present
Human Value Integrated Agent using the Schwartz Value Framework. A value-aware AI system that predicts human values from text using MiniLM, DeBERTa, and RoBERTa, makes decisions through configurable agent profiles and generates natural-language policy justifications using Mistral 7B. Built on the Touché ValueEval Human Values Dataset.
View ProjectCTF_STU032
November 26, 2025 – November 26, 2025
CTF Challenge: Book Review Authenticity Analysis using Machine Learning, SHAP explainability, and data forensics to detect manipulated reviews in e-commerce datasets
View ProjectAI-Code-Reviewer
November 22, 2025 – November 22, 2025
__pycache__/ - Python bytecode cache (regenerates automatically) review_history.json - If you want to start fresh with no history
View ProjectAgricultural-Field-Productivity-Prediction
September 19, 2025 – September 19, 2025
Built ML pipeline to classify fields as low, medium,high productivity using structured agricultural data. Trained,tuned LightGBM achieving 60.5% accuracy ,0.371 F1-score, improving over Random Forest baseline (60.3% accuracy, 0.309 F1). Implemented data preprocessing, validation,automated submission pipeline.
View ProjectPodcast-Listening-Time-Prediction-
July 11, 2025 – July 11, 2025
Implemented regression pipeline to predict podcast listening time based on metadata such as podcast category, duration, language, user engagement patterns. Performed comprehensive data preprocessing including handling missing values, encoding categorical features, feature scaling. Trained using XGBoost, LightGBM, optimizing for MSE.
View ProjectHuman-Activity-Recognition-using-PIR-Sensors
July 11, 2025 – July 11, 2025
Developed a stacked LSTM-based deep learning model to classify human occupancy from PIRvision sensor time-series data. Performed feature scaling , label mapping, random oversampling, and stratified sampling to handle class imbalance and improve model learning. Conducted thorough EDA to visualize sensor activation patterns , class distributions.
View ProjectIMDB_api
July 11, 2025 – July 11, 2025
IMDB_API extracts real-time movie data from IMDB through web scraping, allowing users to search for movie details like ratings and actors. The information is saved in a database and accessed through a RESTful API. It does not utilize machine learning for personalized recommendations or natural language processing.
View ProjectEmail-Sender-Web-Application-
July 14, 2022 – July 14, 2022
Email Sender is a Web Application by which we can directly send the mail to recipient form Our Application. After getting this application we would not be required to go on any of existing mail portal like gmail, yahoo, etc..
View ProjectCultural Fit Analysis
The candidate has a strong portfolio of personal projects, indicating initiative and a passion for data science. The projects cover a good breadth of topics within machine learning and AI, aligning well with an innovative and research-oriented culture. However, the lack of team-based projects or professional experience makes it difficult to fully assess collaboration and adaptability in a corporate environment.
Soft Skills & Operational Fit
The candidate's project descriptions indicate a proactive and hands-on approach to problem-solving. The diversity of personal projects suggests self-motivation and a willingness to explore different areas within data science. However, without specific psychometric test results or interview data, it is difficult to assess soft skills like teamwork, communication, or stress handling.