Hardik Gulati
Second-year B.Tech student in AI & Machine Learning at Symbiosis Institute of Technology. I build data-driven systems — from self-auditing healthcare AI pipelines to fraud detection engines and predictive models. Focused on turning messy data into decisions that matter.
Projects
Self-auditing radiology report generation system. Image-to-text pipeline using Vision Transformer (ViT) + DistilGPT-2, layered with FAISS-based RAG to reduce hallucinations. Evaluated across 7,000+ real medical image-report records using BLEU, ROUGE-L, and BERTScore.
Present
Banking transaction analyser and fraud detection engine. Isolation Forest and Autoencoder-based anomaly detection with SMOTE and cost-sensitive learning to handle severe class imbalance. Evaluated using Precision, Recall, and F1-score.
Smart donation rerouting platform for orphanages — intelligently redirects contributions to where they're needed most. Full backend with Flask and SQLAlchemy: authentication, role-based access, MySQL schema design, and analytics APIs.
Nov 2025
ML model predicting ride-hailing fares from trip features — distance, time of day, traffic patterns, and location data. Regression pipelines, feature engineering, and EDA-driven preprocessing.
Time-series model forecasting inflation trends using macroeconomic indicators. Statistical modelling and deep EDA on CPI movement drivers, with visualisations built for clarity and interpretability.
Skills
Education & Credentials
Open to data & AI
roles & internships.
Looking for opportunities to work on real problems with real data. Let's build something meaningful.