Skip to content
PJ Dailey Tutor | AI Consultant | Content Creator

Projects

Current public projects on GitHub

A focused portfolio of public work across educational security tooling, privacy-conscious vision systems, PyTorch experimentation, and protocol engineering.

Selected Work

Selected builds that show how I approach security, AI, and systems work

These are the public projects that best represent the technical side of the portfolio right now, with a bias toward clear scope, strong engineering fundamentals, and honest framing.

Educational security tooling Security Python

RFID Security Toolkit

Authorized-lab tooling for studying legacy physical-access weaknesses, with Wiegand decoding, MIFARE Classic analysis, and safety-bounded lab guidance.

Python CLI tooling Raspberry Pi GPIO RFID / access control
  • Decodes 26-bit Wiegand frames with parity checks and optional Raspberry Pi GPIO capture.
  • Parses MIFARE Classic dumps, validates access bytes, and flags default trailer keys from a dictionary.
  • Pairs code with lab-safe scenario docs and mitigation guidance for defenders and instructors.
Privacy-conscious computer vision CV Privacy

Face Security Pipeline

On-device identity verification pipeline with YOLOv8 face detection, embeddings, optional anti-spoofing, and FastAPI delivery.

Python FastAPI YOLOv8 facenet-pytorch ChromaDB
  • Keeps biometric embeddings local and exposes metadata only, not raw vectors, to clients.
  • Supports alignment, cosine-similarity identity verification, optional anti-spoof scoring, and MJPEG/WebSocket streaming.
  • Includes optional PostgreSQL attendance logging while preserving a privacy-first default posture.
PyTorch vision toolkit ML PyTorch

Neural Style Transfer Toolkit

Production-oriented neural style transfer toolkit spanning classical Gatys optimization, fast AdaIN, and temporally stabilized video stylization.

Python PyTorch Streamlit Docker Next.js project hub
  • Implements multiple stylization modes including Gatys, AdaIN, and video pipelines with temporal smoothing.
  • Ships CLI entrypoints, a Streamlit demo, and Docker support for reproducible experimentation.
  • Separates the heavy vision toolkit from a lightweight Next.js project hub used for docs and presentation.
Protocol engineering demo Systems Python

bitcoin-node-pure

Learning-oriented Bitcoin codebase that keeps core cryptography and transaction logic stdlib-only while layering a friendlier CLI on top.

Python secp256k1 primitives P2P networking SQLite cache
  • Covers ECC, hashes, legacy transactions, and P2P framing under a clean package structure.
  • Adds a richer CLI with environment checks, key generation, tx encoding flows, and peer diagnostics.
  • Positions scope honestly as an engineering demonstration rather than a full consensus node.
GitHub logo

More repositories

Want the full repository list?

The selected cards above are the most portfolio-relevant public projects, but the full GitHub profile includes more experiments and supporting repos.