Projects in KangraProjects in Kangra
Projects
People
Cover image for 1. What it is: An
1. What it is: An advanced e-commerce web app that combines Graph Neural Networks (GNNs) and Generative AI to identify "toxic inventory"—products, sizes, or suppliers driving high customer returns. 2. GNN Predictive Modeling: Built a heterogeneous GNN using PyTorch Geometric to model relationships between customer demographics and product features to predict future return risks. 3. AI Attribute Enrichment: Automated a metadata extraction pipeline using the Gemini API with local caching to parse raw HTML product descriptions into structured product features (fabric, fit, pattern). 4. Interactive Dashboard & Reporting: Developed a Streamlit dashboard with Supabase Auth and integrated ReportLab to generate boardroom-ready PDF return audit reports. 5. Tech Stack: Python, PyTorch Geometric, Streamlit, Gemini API, Supabase, ReportLab.
0
21
Cover image for This project showcases a premium
This project showcases a premium presentation layout highlighting a series of high-conversion Social Media Creatives. The design approach is engineered for direct-to-consumer (DTC) e-commerce brands in the wellness, beauty, and luxury lifestyle sectors, focusing on a seamless blend of elevated aesthetics and strategic marketing. Design Strategy & Layout Highlights Typography Hierarchy: The main presentation utilizes an elegant, high-contrast serif typeface for the headers ("SOCIAL MEDIA CREATIVES"), paired with a fluid, organic script accent for a touch of luxury. The ad creatives use clean, accessible sans-serif typography designed for maximum readability on mobile feeds. Color Story & Tone: A deeply cohesive, warm neutral color palette dominates the canvas—utilizing soft cream, warm beige, muted sage green, and striking contrast accents of deep charcoal and rich espresso brown. Multi-Product Mockup Integration: As showcased in the image, the composition features an array of modular square grid layouts alongside a central smartphone UI mockup. This setup effectively demonstrates how the campaign assets transition seamlessly from Instagram grid posts to high-impact mobile stories or reels. Visual Styling: The entire layout leverages soft, natural leaf and window-pane shadow overlays, raw stone styling elements, and warm, minimalist photography to communicate a sense of "quiet luxury" and organic purity.
0
31
Cover image for Scientific Image Forgery Detection — Kaggle Competition Part...
Scientific Image Forgery Detection — Kaggle Competition Participated in the ongoing Kaggle competition on Copy-Move Forgery Detection in Scientific Images, aimed at identifying manipulated biomedical figures that can compromise research integrity. For this challenge, I developed a ResNet50 + U-Net hybrid segmentation model using PyTorch, designed to detect and segment forged regions at the pixel level. My approach combines Dice and Focal losses for balanced training, WeightedRandomSampling to oversample forged images, and Test-Time Augmentation (TTA) to improve prediction robustness. Achieved an initial score of 0.303 on the public leaderboard. I’m continuing to experiment with architecture tuning, learning rate schedules, and other loss functions to further enhance performance and generalization.
0
87