RUNTIME

--:--

Say hello

All projects

Each one shows what it is, the key facts, the stack, an honest status and every link. Open a case study for the problem, the approach and the key engineering decision.

  1. Screenshot of Wardlock's live site

    PayPal AI Hackathon 2026

    LiveNow showing

    Wardlock

    Safe PayPal refunds for AI support agents

    A gateway that lets an AI support agent refund PayPal customers on its own, without being able to refund the wrong person, the wrong amount, or behind the merchant's back.

    • The agent's refund tool has no amount or payment ID
    • 38 scripted attacks, 0 allowed
    • Solo, built for the PayPal AI Hackathon 2026
    • Python
    • FastAPI
    • MCP
    • PayPal REST APIs
    • SQLAlchemy
    • PostgreSQL
    • pytest
    • React
    • Vite
    • AG Grid
    • Render
    Case study Live demo Source code Video

    The demo runs on Render's free plan and sleeps when idle, so the first load after a pause takes 50–60 seconds. Open the console and choose View the live demo (read-only).

  2. Screenshot of H-Tech's live site

    Solo project

    LiveNow showing

    H-Tech

    Hybrid movie recommender

    A movie recommender that blends content-based and collaborative filtering, trained on MovieLens 25M and enriched with TMDB posters and metadata.

    • Live, deployed free on Cloud Run, Supabase and Vercel
    • Trained on MovieLens 25M ratings
    • Solo, end to end
    • Python
    • scikit-learn
    • Polars
    • ONNX Runtime
    • FastAPI
    • SQLAlchemy
    • PostgreSQL
    • Supabase
    • Next.js
    • TypeScript
    • Docker
    • Google Cloud Run
    • Vercel
    • GitHub Actions
    Case study Live demo Private repo · code walkthrough on request

    The first load can take 30–60 seconds. The backend runs on Google Cloud Run's free tier and needs a moment to wake up.

  3. Engineering Clinics, VIT-AP

    Prototype readyRough cut

    Smart Vision Screening Device

    Edge-AI eye screening

    A low-cost, portable eye-screening device. It photographs a patient's eye, runs a machine-learning model on the device itself, and sends the result to a web dashboard where a doctor reviews it and issues a prescription.

    • YOLOv8-Nano, INT8-quantised to ~1.8 MB, runs on the device
    • Works offline in screening camps; results upload later
    • Dashboard in English, Hindi and Telugu
    • C++
    • Arduino
    • TensorFlow Lite for Microcontrollers
    • YOLOv8-Nano
    • XIAO ESP32S3 Sense
    • HTML
    • CSS
    • JavaScript
    • Firebase
    • Node.js
    • Inline SVG charts
    Case study Private repo · code walkthrough on request
  4. In progressIn production

    Java HNSW Vector Engine

    A vector search engine in Java, built on HNSW graphs.

    • Java