Native C++20 Linux CLI tool built to recover lost worlds due to Early Access game bugs and save lockouts.
When a player loses a Hardcore world to an unfair glitch (e.g., getting hit through walls or geometry bugs),
the in-game setting permanently locks itself. This tool performs binary reverse engineering on PlM1 (Oodle/Kraken) and PlZ (zlib)
save containers, decompressing and editing the internal bHardcore & bPalLost flags to restore or re-synchronize
permadeath status. Features automatic detection of Steam, Steam Deck, Proton, and Flatpak save paths, safety backups before editing,
and JSON mode for scripting.
C++20
Reverse Engineering
Oodle / Kraken
zlib
Linux / Steam Deck
Open Source
Proton / Steam
Competitive desktop app (PySide6/Qt) designed to help newcomer players build competitive teams with their favorite Pokémon to beat the meta.
Its core feature is an Interactive Manual Team Builder: players search Pokémon from a 315-species catalog and manually configure custom builds (EV/IV distributions with 66-point allocations, valid movepool selections, items, and abilities) with usage trend guidance. Includes dynamic Mega-Evolution toggling (75 forms) recalculating base stats in real time.
The app runs a deep Synergy & Gap Analysis Engine that diagnoses team weaknesses, coverage flaws, and stat synergies. At match start, the Team Preview Assistant uses screen capture & sprite matching (OpenCV) to identify the opponent's roster and instruct which Pokémon from the user's team to bring into battle.
Status: The Team Builder, Synergy Analyzer, Mega-Evolution engine, movepool validator, and SQLite database (+390 moves/stats) are fully functional and backed by 449 unit tests and a GitHub Actions CI pipeline. Real-time in-game battle HUD recommendations are under active development.
Python
PySide6 / Qt6
SQLite
Synergy Analyzer
OpenCV / Sprite Matching
CI / GitHub Actions
Unit Testing (449 Tests)
End-to-end IoT system built on Arduino with a YL-69 moisture sensor
and water pump. Features automatic control via configurable thresholds, manual override,
a Python GUI with 24h history chart, and a local REST API on port 8080 with a
mobile-responsive web panel — all running 100% offline.
Python
Arduino C++
Tkinter
PySerial
SQLite
IoT
Open Source
REST API
Automated security system with motion sensors, automatic door control,
audible/visual alarm and surveillance camera. Includes nighttime logic (01:00–05:00)
with autonomous activation. Runs on Raspberry Pi, Cisco Packet Tracer, or any PC
in simulation mode — same codebase, zero changes.
Python
Raspberry Pi
GPIO
Packet Tracer
IoT
Open Source
Full penetration test of a deliberately vulnerable Flask app:
five findings — Command Injection, Reflected XSS, CSRF, SQL Injection and an
exposed debug console (RCE) — each located in source, proven with an
OWASP ZAP scan of 999 requests plus manual exploits, and shipped
with its own remediation. Every attack was run and verified to fire against the
target and to be blocked against the hardened version.
OWASP ZAP
Burp Suite
OWASP Top 10
Python
Flask
SQLite
Open Source
What I'm building next: an AI-assisted honeypot that maps
attacker TTPs to MITRE ATT&CK, and a LoRa + MQTT IoT mesh for
connectivity-free environmental monitoring. Repos go public as they take shape.
Python
Machine Learning
MITRE ATT&CK
LoRa
MQTT
IoT Security
