Project Showcase

Live full-stack DevSecOps platform featuring real-time OWASP Top 10 API vulnerability auditing, cryptographic JWT token inspection, and Supabase Row-Level Security (RLS) policy simulation.

Problem

Modern Next.js & Supabase applications suffer from silent security misconfigurations, unparameterized query injections, broken object-level authorization (BOLA), and insecure JWT verification defaults that allow unauthorized multi-tenant data access.

Outcome

Delivered an interactive, production-grade security auditing suite that identifies critical vulnerabilities in real-time and generates hardened Next.js / Supabase code fixes with zero false positives.

Interactive analytics suite for event telemetry, threat detection logs, and real-time visualization.

Problem

Teams tracked numbers across spreadsheets and disconnected dashboards, with no shared tenant model or consistent UI patterns, leading to authorization vulnerabilities and data fragmentation.

Outcome

Delivered a working multi-tenant dashboard prototype with role-aware layouts, reusable chart widgets, and strict security isolation ready for real data integration.

Headless storefront focused on product storytelling and a clean mobile checkout path.

Problem

Product pages felt slow, loading states were jarring on slow mobile connections, and simple content merchandising updates required developer intervention, resulting in high churn.

Outcome

Shipped a responsive, content-friendly commerce frontend featuring instant transitions, optimized mobile product pages, and Stripe integration.

Real-time SSE messaging platform with end-to-end payload sanitization and RBAC session security.

Problem

Support teams needed an in-product assistant surface that felt native and snappy, but existing iframe embeds had poor accessibility, broke mobile layouts, and suffered from unpredictable latency.

Outcome

Built a lightweight, embeddable support chat widget with sub-50ms perceived latency, live typing indicators, and reliable SSE streaming.

Final Year Project — A clinical web application that predicts patient blood glucose levels using a trained ElasticNet ML model and provides a real-time daily health monitoring dashboard.

Problem

Medical staff lacked a fast, data-driven tool to predict blood glucose levels from patient vitals. Manual tracking across spreadsheets was error-prone and provided no early alert for high-risk glucose readings.

Outcome

Delivered a fully functioning clinical prediction system with ML-powered glucose forecasting, admin and staff role access, automated high-risk alert flags, and exportable patient reports.

Lightweight internal console for deploy health and incident visibility during handoffs.

Problem

Status lived in fragmented chat threads and log consoles; on-call engineers needed one calm, centralized place to scan service states and incident histories during shifts.

Outcome

Delivered a focused status console layout that reduces context-switching during on-call handoffs and provides sub-50ms query latency.