Back to Home

Projects

Signal-driven engineering. Factual and precise.

FeedforwardNN

C++PythonMake

A feedforward neural network implemented from scratch in C++ to demonstrate the math behind forward and backward passes.

Technical Details

  • Architecture: 2 input → 1 hidden layer (configurable) → 1 output
  • Activation: Sigmoid (hidden + output)
  • No bias terms
  • Weight initialization: Random
  • Training: Online / stochastic gradient descent
  • Loss: Absolute error (averaged, logged every 100 epochs)
  • Training length: 10,000 epochs
  • Learning rate: 0.1
  • Logging: CSV output + Python loss plot
  • Purpose: clarity over production features
Tone: mathematical, instructional, honestView Source

LanHUB

Next.jsTypeScriptWebCrypto API

A LAN-focused, end-to-end encrypted collaboration app built with Next.js and TypeScript.

Engineering Emphasis

  • Client-side cryptography
  • Per-recipient key packing
  • Server never sees plaintext or raw symmetric keys
  • In-memory server + heartbeat polling
  • LAN discovery without heavy infra
  • LocalStorage persistence
  • Explicit client-side key management
Tone: security-aware, system-design focusedView Source

TrickStroke

Socket.IONode.jsReact

A real-time social-deduction word game using Socket.IO with authoritative server logic.

System Flow

  • Join flow
  • Game start
  • Role assignment
  • Word submission validation
  • Real-time chat
  • Voting and leaderboard updates
  • Server-authoritative state
Tone: event-driven, real-time systems thinkingView Source