Project Terminal

LLM fine-tuning project: Qwen2.5-3B-Instruct + QLoRA transforms vague prompts into structured, effective ones. Full pipeline (data generation, training, evaluation, HuggingFace publish)

LLM security testing framework with jailbreak detection, defense testing, and JailbreakBench integration

Prompt engineering toolkit with Prompt Doctor analysis and pattern library

Verified software-agent evaluation: challenge tasks, trajectory annotation, and scalability analysis for LLM agents

PowerShell tool for diagnosing VDI performance issues

Tool for network design and planning

Meta honeypot implementation for security research

GUI-based backup solution with Tkinter

Compression algorithms and testing

Multi-factor authentication implementations

Collection of penetration testing utilities

Web-based hash generation tool

Polymorphic malware detection demonstration

Tool for detecting suspected lagswitch behavior