~/portfolio $ whoami
Full-stack developer working across East and West Coast time zones. I build things that run in production — not demos, not prototypes.
Currently deep into AI tooling: local LLM fine-tuning pipelines, browser extensions, and the intersection where backend systems meet language models.
| Laravel / PHP | Primary | |
| React | Primary | |
| MySQL | Primary | |
| LLM / MLX | Growing | |
| Node / JS | Fluent | |
| Python | Fluent |
End-to-end pipeline: export conversation data from multiple AI platforms, score, filter, clean, desensitize, convert, and fine-tune. Architecture insight: scoring before desensitization prevents placeholder leakage. Working adapter: local_claude_final2.
Browser extension injecting branch-marking UI into Claude, Gemini, ChatGPT, and Grok simultaneously. Platform-specific adapters handle DOM differences across four distinct interfaces. Zero dependencies.
Laravel queue architecture with timezone-aware scheduling. Debugged UTC date-shifting bugs in JavaScript ISO string interpretation, MySQL charset/collation conflicts, and Eloquent belongsToMany edge cases in production.
Two-stage classification: Claude API (Haiku) for topic judgment, GPT for format summarization. Each model used where it outperforms the other. Designed to run before data desensitization to prevent placeholder contamination in training sets.
Break tasks by complexity before touching code. Know the blast radius before you dig.
File by file. Line by line. Fresh context when things go in circles. No spaghetti.
Production first, polish second. Things that run matter more than things that look good in dev.
CHENGHONG_MENG_UNSTOPPABLE_FOUNDATION
LLAMA_3_PERSONALITY_SYNC_COMPLETE
TARGET_LOCKED: NEXT_OPPORTUNITY
You found it. Now hire me.