blog
◐
☀
☾
updated 2026-06-20
All SIDE PROJECT RESEARCH PROJECT HACKATHON PROJECT AI Agent PM AI Builder Website
SHIP-01
⋮⋮
Lark Loom
The collaborative AI Agent rooted in the Lark ecosystem.
HACKATHON PROJECT AI Agent PM Delivery Layer Developer Collaboration Lark Lark AI Challenge for College Students
SHIP-02
⋮⋮
Citation Trap
Agentic RAG Reference Benchmark for Trustworthiness Evaluation
HACKATHON PROJECT AI Agent PM Benchmark
SHIP-03
⋮⋮
Gmail Plus Plus
Gmail++ is a AI-powered email ranking platform that gives you a read-only reply queue on top of Gmail, with visible reasoning and reversible account-scoped preferences.
SIDE PROJECT AI Builder XGBoost Productivity Website
SHIP-04
⋮⋮
Worth Fly
WorthFly currently focuses on search, decision support, alert drafts, and external purchase handoff. Final fare, inventory, and ticketing still belong to the real booking channel.
SIDE PROJECT AI PM 0→1 Explainable AI Website
SHIP-05
⋮⋮
SusBench
Online benchmark for evaluating the susceptibility of CUAs to UI dark patterns.
RESEARCH PROJECT HCI IUI 2025
NOW-01
⋮⋮
Landed
1
SIDE PROJECT
NOW-02
⋮⋮
YiTang Profile Matching Infra
1
INTERN PROJECT
NOW-03
⋮⋮
ApplyMint
ApplyMint is a AI-powered chrome extension that compiles the resume into JSON, import it once for visual editing, and automatically fill it in one click on the job application form
SIDE PROJECT Job Application Chrome Extension
NOW-04
⋮⋮
SleepStreak
Designing Healthier Pre-Sleep Phone Use Experience for Teens & Early Adults
RESEARCH PROJECT
NOW-05
⋮⋮
SusFix
Ongoing project about building a system about using LLM to auto fix the dark patterns on the websites.
RESEARCH PROJECT
NEXT-01
⋮⋮
From PM to Builder
Stop briefing engineers — become one. Use solo builder loops to solve real business problems end-to-end, not just spec them.
Career Pivot AI Builder 0→1
NEXT-02
⋮⋮
Make AI worth looking at
Most AI products inherit the model's flat default aesthetic. Explore distillation + skill composition to ship AI that's well-crafted, not just functional.
AI Aesthetic Distillation Skills Craft
NEXT-03
⋮⋮
Close the product loop
Build operations + data analytics fluency so I own the full product lifecycle — not just the design-and-ship slice.
Product Ops Data Analysis Full-stack PM
LATER-01
⋮⋮
AI for people, not just AI people
The 1-3 year version of me: shipped AI products that ordinary people — not the early-adopter or AI-insider crowd — use daily to make their lives meaningfully better.
Mission Mass-impact AI Real-world AI
LATER-02
⋮⋮
Human-centered, agent-aware
Human-centered design stays the foundation. But in the agent era, AI is also a user — and most products still ignore that. The 1-3 year version of me: deep enough AI fluency to design for both humans and the agents acting on their behalf, so the AI actually helps.
Human-centered Design Agent-friendly HCI × AI Pi-shaped
$ ./agent
terminal
type help · ↑↓ history
$ ls ~/skills
/skills
dotfiles for AI — what I built to multiply my own AI work
SKILL-01
/devloop
Claude Code skill
autonomous unattended dev loop — branch → small step → test → PR → sub-agent review → merge → verify deploy → next
SKILL-02
/ai-pm-resume
Claude Code skill
AI PM résumé reviewer — 8-dim signal audit + 5-reader critique + anti-fab guardrails + tier adaptation
SKILL-03
/interview-sim
Claude Code skill
synthetic interview generator — protocol + persona → time-budgeted transcript that reads like recording transcription, not LLM output
SKILL-04
/log
Claude Code Skill
When you use Claude Code for projects, it helps you keep a diary in the background. When the project ends, this diary can be easily turned into a blog draft with one click.
SKILL-05
Claude Credit Watchdog
Claude Code Plugin
A lightweight menu bar watchdog that monitors your Claude Code sessions, alerts you the moment credits renew, and automatically resumes your work.
SKILL-06
agent-brief.txt
agent infra
off-page notes my site's QA Worker reads — context I want an agent to know but not render on the page
$ grep -i
how_i_think.md
principles · updated occasionally
/ 01
Evals before features.If you can't measure it, you're shipping vibes.
/ 02
The model is one input.Latency, UX, and trust often matter more than benchmark deltas.
/ 03
Demos lie. Production tells the truth.Watch the long tail.
/ 04
A great PM removes ambiguity. Models add it back.The job is balance.
$ /agents
how this site works for AI
three entry points
This site is designed to be read by both humans and AI agents. The dashboard, terminal, and detail panels are for you. For an agent, prefer:
Commands
⌘K