Hu Qi avatar
Hu Qi
huqi.host
deploying to https://huqi.host

Building useful things with AI — indexed, refreshed, and useful.

AI-native builder · Front-End engineer · Cloud & AIoT explorer. I build practical AI applications, developer tools, technical tutorials, and cloud-native demos. Recently focused on AI Agent, RAG, full-stack engineering, ROS2, and CloudDev.

production.profile / refreshed: 2026-05-26T12:01:58.968Z
Hu Qi

Hu Qi

@hu-qi · Canton

Finding myself. Building useful things with AI.

131
GitHub repos
GitHub API
28
Stars
latest public repos page
123
Followers
GitHub API
675
Following
GitHub API
knowledge map

Focus as an operating system

Not a skill list, but a navigable knowledge system across AI, front-end, cloud, and technical writing.

deep

AI Agent

from toy demos to useful workflows

repos · tutorials · cases

deep

RAG

retrieval quality, knowledge engineering, evaluation

notes · examples · templates

build

Front-End

interaction, visual polish, developer experience

components · pages · docs

cloud

CloudDev

practical cloud scenarios and hands-on paths

case studies · guides

edge

AIoT / ROS2

bridging software and devices

experiments · demos

write

Technical Writing

turning exploration into reusable knowledge

blogs · WeChat · open-source notes

public signals

Projects, communities, trails

Project descriptions use public signals only and avoid unverifiable resume claims.

Filter
seo / geo

Search and LLM-readable by default

SEO is not keyword stuffing. It is about making the profile understandable, useful, sourced, and easy to contact.

Production SEO checklist

Use https://huqi.host as canonical origin everywhere.
Keep primary content in indexable text, not only animated client-only UI.
Add ProfilePage JSON-LD for person identity and sameAs links.
Ship sitemap.xml and robots.txt; submit the domain property in Search Console.
Use descriptive sections and stable source links.
Avoid thin AI-generated pages and unverified resume claims.
ai layer

Profile Copilot, backed by server API

The production version calls the LLM from /api/chat. If OPENAI_API_KEY is missing, it falls back to deterministic local answers.

Ask from profile data

Try: SEO / LLM / Vercel / refresh / RAG

Hi, I am Profile Copilot.

Ask me about AI/RAG/front-end/CloudDev, or the SEO, LLM, Vercel, and refresh architecture behind huqi.host.

LLM integration guardrails

Never expose OPENAI_API_KEY or GITHUB_TOKEN in the browser.
Use /api/chat as a server-side proxy with local fallback.
Ground answers in the profile payload refreshed by /api/cron/refresh.
Return source labels for GitHub, blog, and community links.
Say when a claim is not supported by the current profile context.
contact

Ship it at huqi.host.

AI applications, front-end, CloudDev, AIoT, tutorials, and open-source learning projects.

Hu Qi | AI-native Builder, Front-End Engineer, Cloud & AIoT Explorer