Illustrated portrait of Xinwei Xiong

Xinwei Xiong · Practitioner and writer Building systems and understanding people

Turning lived practice
into knowledge that stays useful.

I build AI systems and real products, document the workflows and engineering judgment behind them, and write about growth, philosophy, and life.

Xinwei travelling and building in the field
GET CLOSE Judgment needs contact with reality.
01 / 03

A map of everything I have made public.

Projects, working methods, personal reflections, and annual reviews, organized by subject and time.

Annual reviews

A yearly check on the direction of life.

Recent monthly notes

Six months, six fields of thought.
All monthly notes

LONG-TERM THREADS

Three connected worlds.

01 / 03
Telepace product interface
AI and projects

Turn deep workflows into real, testable systems

Openly documenting project process, AI collaboration, code, and key tradeoffs.

Products in motion, not portfolio decoration.

Each began with a problem I faced and keeps changing in the hands of real users.

Open the full product workspace →
01 / 06

01 / 11

The road is not a break from life.
It is life itself.

Twelve months, nine countries, more than 35 cities, and over 400 km on foot. Travel keeps undoing my untested answers.

Explore the travel map
Travel photograph from South Taihang, China
South Taihang, China

OPEN INTERVIEW / ABOUT 4 MIN

Show me what I have not seen.

Talk with my AI interviewer for four minutes. Your answers become evidence for the next iteration.

Start with one concrete reaction

Anonymous, leave at any time

The live study is being prepared.

Building AI products, user research, or open-source tools?

Bring one concrete problem.

Want to read my writing? Open four curated reading paths

Now that you know the person, meet the writing. Everything on this blog follows four threads: AI agents, engineering & open source, growth & thinking, and travel & observation. Not sure where to begin? Here is the map drawn for your first visit.

Just want one post? Wandering & Growing: 2025-2026 Annual Review A year in deep review — the tech, the growth, the travel, all in one place. Read this and you will roughly know me. 25 min read Start with this one

Choose your path

Four threads, a few representative posts each. Pick the one that pulls you in.

AI Agent

For developers curious about LLM applications, agent engineering, and open-source AI deep-dives — the highest-traffic thread on this site.

  1. Microsoft MarkItDown 0.1.6: A Practical Document-to-Markdown Guide A deep read of Microsoft's MarkItDown — the most-searched page on this site. 15 min read
  2. LangGraph Architecture in 2026: StateGraph, Persistence, and Recovery LangGraph's design, architecture, and principles, dissected. 7 min read
  3. Gemini Notebook (Formerly NotebookLM): A Source-Grounded Research Workflow The RAG engineering behind Google NotebookLM. 18 min read
  4. Agent Identity: From Locke to OpenClaw From Locke's personal identity to the identity of AI agents — a signature essay. 16 min read
  5. Mem0 OSS v3 in Practice: Memory Architecture, Retrieval, and Trade-offs Giving AI a memory: the mem0 architecture, unpacked. 11 min read
Browse all AI Agent posts

Engineering & Open Source

For engineers who care about Kubernetes, cloud native, Go, and open-source practice.

  1. About My Hugo teaching How this blog is built: a complete Hugo walkthrough. 15 min read
  2. Kubernetes an Article to Get Started Quickly The evergreen zero-to-one introduction to Kubernetes. 24 min read
  3. GoReleaser v2 + GitHub Actions: Go Release Guide Go release tooling and engineering practice, end to end. 53 min read
  4. The Art of Asking Questions in Open Source Communities The art of asking good questions — lesson one of open source. 48 min read
Browse all engineering posts

Growth & Thinking

For anyone on the road of self-discovery, annual reviews, and sharper thinking.

  1. Maintaining Self-Worth in the Age of AI When AI can do everything, where does the weight of the self sit? 10 min read
  2. Flow State: Deep Focus and Happiness Guide How to enter and hold a flow state — a complete guide. 3 min read
  3. How I Learned to Observe My Own Thinking: A Four-Month Metacognitive Shift How I learned to observe my own thinking over a four-month metacognitive shift. 11 min read
  4. GTD and the Quadrant Method Practice GTD and the four-quadrant rule, as a system that actually runs. 13 min read
  5. The Super-Individual Stack: AI-Native Product Directions and Solo Builder Ops in 2026 The super individual: how far can one person plus AI go? 21 min read
Browse all growth & thinking posts

Travel & Observation

For digital nomads, hikers, and cultural observers.

  1. Travel · The World Game An interactive footprint map across 9 countries — the whole journey at a glance. Interactive map
  2. In 2023, I Was Wandering at the Edge of the World A year of wandering at the edge of the world — digital nomad life, unfiltered. 36 min read
  3. Japan Travel Notes — Learning to Be with Time Through Wood, Fire, and Gaps Summiting Mount Fuji, and the rest of Japan. 14 min read
  4. Lhasa: Slow and Heavy Lhasa: slow and heavy — field notes from the plateau. 1 min read
See the full travel log

About this site

cubxxw.com is Xinwei Xiong's open archive of practice and thought. Its writing grows from real projects, engineering choices, and lived experience, covering AI workflows, agent engineering, open-source practice, systems thinking, personal growth, philosophy, and life. Xinwei has contributed to projects including OpenIM, OpenKF, and Sealos, and openly documents the methods, failures, and judgment formed through practice.

Author
Xinwei Xiong (cubxxw)
Identity
Practitioner · Writer · Open-source contributor
Threads
AI systems / Engineering methods / Real projects / Growth & philosophy
Footprint
9 countries · 35+ cities · 400+ km hiked