ChiChieh HuangFOUNDER · AI ENGINEER
A map of my writing · 2023.12 — 2026.11

Three years of writing,
one cairn at a time.

107 essays and 105,214 words on AI, written first in Traditional Chinese and translated here. Each essay is a cairn on this map. The trail runs from the Data Lowlands of 2023 up to Reflections Meadow today. Hover a cairn to see what it’s about.

Read →
Cairn = one essay; more stones, longer essay
Monthly writing trail
Contours = topics building up
3 months
Where I am now · Latest trail

The latest trail, Agent Memory

Season one ran four essays, from whether we even mean the same thing by “memory” to how agent memory forms, updates and fades. Season two asks whether an agent that remembers also remembers right.

Season 1 · Keeping the past
S1·1

Everyone Talks About Memory. Do We Mean the Same Thing?

Everyone working on AI agents talks about memory, but chat history, MEMORY.md, RAG, context engineering and fine-tuning answer different questions. The first essay in my Agent Memory series pulls them apart.

1,881 words2026.08.24
S1·2

What Does an Agent's Memory Actually Look Like?

Where an agent's memory lives and what form it takes are two different questions. A tour of token-level, parametric and latent memory, and where MEMORY.md, vector DBs and knowledge graphs fit.

2,601 words2026.08.15
S1·3

What Should an Agent Remember?

Not everything that happens is worth an agent remembering. Factual, experiential and working memory each do a different job, and stacking them against memory's form gives you a map of the field.

3,027 words2026.08.24
S1·4

How Agent Memory Forms, Updates and Fades

Memory isn't a static snapshot. It has a lifecycle of formation, evolution and retrieval, with consolidation, updating and forgetting in between. The season one finale of my Agent Memory series.

3,230 words2026.08.27
Season 2 · When the past stops applying
S2·1

Remembering Isn't the Same as Remembering Right

Once an agent stops forgetting, the next problem is whether it keeps remembering wrong. How to build reliability into writing, updating and reading memory, and how to test that it actually worked.

3,936 words2026.09.16
S2·2

When Memory Gets Attacked

Memory poisoning turns an agent's own notes against it. A made-up discount case shows how a claim becomes "policy," and how to separate notes from rules, gate actions, clean up and test.

2,767 words2026.10.12
S2·3

Can an Agent Learn to Manage Its Own Memory?

Remembering correctly isn't enough if the agent can't pick, tidy and weigh its notes. What recent research says about learned memory policies, and how to test them against the next real task.

3,594 words2026.10.28
Words so far21,036
010,00020,000S1·1S1·2S1·3S1·4S2·1S2·2S2·3S1·1 Everyone Talks About Memory. Do We Mean the Same Thing? · 1,881 wordsS1·2 What Does an Agent's Memory Actually Look Like? · 4,482 wordsS1·3 What Should an Agent Remember? · 7,509 wordsS1·4 How Agent Memory Forms, Updates and Fades · 10,739 wordsS2·1 Remembering Isn't the Same as Remembering Right · 14,675 wordsS2·2 When Memory Gets Attacked · 17,442 wordsS2·3 Can an Agent Learn to Manage Its Own Memory? · 21,036 words
Field log · All essays

107 essays, by region

Regions are the place names on the map above; pick one and its cairns light up.

2026.11.01 Reflections Meadow Speaking at AI Impact Day in the Data Village 210 words 2026.10.28 Memory Summit Can an Agent Learn to Manage Its Own Memory? 3,594 words 2026.10.19 Reflections Meadow My Old Gaming PCs Are Now CI Runners 756 words 2026.10.16 Reflections Meadow Last Year I Learned to Look Up at the Vision, This Year to Look Down at My Feet 2,813 words 2026.10.12 Memory Summit When Memory Gets Attacked 2,767 words 2026.09.30 LLM Plateau When You Can Take Your ChatGPT Subscription With You, How Does the AI Product Business Change? 886 words 2026.09.23 Memory Summit Jev Lab: Deciding What an Agent Should Remember 436 words 2026.09.16 Memory Summit Remembering Isn't the Same as Remembering Right 3,936 words 2026.08.27 Memory Summit How Agent Memory Forms, Updates and Fades 3,230 words 2026.08.24 Memory Summit Everyone Talks About Memory. Do We Mean the Same Thing? 1,881 words 2026.08.24 Memory Summit What Should an Agent Remember? 3,027 words 2026.08.15 Memory Summit What Does an Agent's Memory Actually Look Like? 2,601 words 2026.08.02 MCP Pass MCP Goes Stateless: Getting Ready for Scale 1,947 words 2026.07.21 Memory Summit AWS Summit × Agile Taipei: Two Packed Talks on Agent Memory 309 words 2026.07.05 MCP Pass Productizing MCP: Where to Draw the Line as Tools Pile Up 2,311 words

All essays →

Footprints · Talks in 2026

32 talks given this year

1 more scheduled.

An ascent profile of talks given this year. Marked points are major events; the dashed line is still ahead.

010203040JanFebMarAprMayJunJulAugSepOctNovDecToday 11.162026-01 GenAI Meetup: The Three-Sword Engineer (Total 1)2026-01 Twinkle AI × GDG anniversary (Total 2)2026-01 R-Ladies Taipei × NTPU Statistics: From Side Project to Business: Building AILogora with Vibe Coding (Total 3)2026-02-11 GenAI Meetup: Raising Your OpenClaw Agent Well (Total 4)2026-02 Internal talk at a tech company: The Do-Everything OpenClaw Assistant (Total 5)2026-02 Internal talk at a sports-tech company: The Do-Everything OpenClaw Assistant (Total 6)2026-02 BRP OpenClaw Workshop (Total 7)2026-03 Taiwan OpenClaw Meetup (MixerBox) (Total 8)2026-03 AppWorks AI Circle (Total 9)2026-03 NACC talk (Total 10)2026-03 GDG study group (Total 11)2026-04 Chung Shan Medical University (Total 13)2026-04 Corporate training series (Total 15)2026-04 GDG on Campus NCCU: Keeping Your Own Judgment When AI Moves This Fast (Total 18)2026-04 Taiwan OpenClaw Meetup, business edition (MixerBox) (Total 19)2026-05-18 AIA × Claude Code (Total 20)WiDS Taiwan2026-05-24 WiDS Taiwan (Women in Data Science): Five AIs on My Team: Building Your Agent Team (Total 21)COMPUTEX / InnoVEX2026-06-04 COMPUTEX / InnoVEX 2026: Open Source Team Taiwan Pavilion (Total 22)AWS Summit Taipei2026-07-16 AWS Summit Taipei 2026: From AI Sessions to Shared Memory: Knowledge Infrastructure for AI-Native Teams (Total 23)Agile Taiwan2026-07-16 Agile Taiwan: From OpenClaw to Cairn: When Agents Join the Team, How Should Memory Be Shared? (Total 24)2026-08-04 AIA Meetup (Total 25)Taiwan AI Annual Conference2026-08-18 Taiwan AI Annual Conference 2026: From AI Sessions to Team Memory: Designing and Building Cairn (Total 26)2026-10-08 AI Tinkerers Taipei — Kickoff: Technical Show-and-Tell: Cairn: Building Reliable Memory for Humans and AI (Total 27)2026-10-18 Corporate training (Total 28)2026-11-01 Data Breakfast × Social Good Special (Data Village): Why Does My AI Act Like It’s Always Its First Day at Work? (Total 29)2026-11-05 AppWorks talk (Total 30)2026-11-11 Generative AI Conference Meetup 2026 (Total 31)2026-11-15 Taiwan AEC Open-Source Skills Knowledge Base Meetup #4: AI Agent Memory and Knowledge Applications (Total 32)2026-12-18 Talk for a startup development association (Total 33) 010203040JanMarMayJulSepNovToday 11.162026-01 GenAI Meetup: The Three-Sword Engineer (Total 1)2026-01 Twinkle AI × GDG anniversary (Total 2)2026-01 R-Ladies Taipei × NTPU Statistics: From Side Project to Business: Building AILogora with Vibe Coding (Total 3)2026-02-11 GenAI Meetup: Raising Your OpenClaw Agent Well (Total 4)2026-02 Internal talk at a tech company: The Do-Everything OpenClaw Assistant (Total 5)2026-02 Internal talk at a sports-tech company: The Do-Everything OpenClaw Assistant (Total 6)2026-02 BRP OpenClaw Workshop (Total 7)2026-03 Taiwan OpenClaw Meetup (MixerBox) (Total 8)2026-03 AppWorks AI Circle (Total 9)2026-03 NACC talk (Total 10)2026-03 GDG study group (Total 11)2026-04 Chung Shan Medical University (Total 13)2026-04 Corporate training series (Total 15)2026-04 GDG on Campus NCCU: Keeping Your Own Judgment When AI Moves This Fast (Total 18)2026-04 Taiwan OpenClaw Meetup, business edition (MixerBox) (Total 19)2026-05-18 AIA × Claude Code (Total 20)WiDS Taiwan2026-05-24 WiDS Taiwan (Women in Data Science): Five AIs on My Team: Building Your Agent Team (Total 21)COMPUTEX / InnoVEX2026-06-04 COMPUTEX / InnoVEX 2026: Open Source Team Taiwan Pavilion (Total 22)AWS Summit Taipei2026-07-16 AWS Summit Taipei 2026: From AI Sessions to Shared Memory: Knowledge Infrastructure for AI-Native Teams (Total 23)Agile Taiwan2026-07-16 Agile Taiwan: From OpenClaw to Cairn: When Agents Join the Team, How Should Memory Be Shared? (Total 24)2026-08-04 AIA Meetup (Total 25)Taiwan AI Annual Conference2026-08-18 Taiwan AI Annual Conference 2026: From AI Sessions to Team Memory: Designing and Building Cairn (Total 26)2026-10-08 AI Tinkerers Taipei — Kickoff: Technical Show-and-Tell: Cairn: Building Reliable Memory for Humans and AI (Total 27)2026-10-18 Corporate training (Total 28)2026-11-01 Data Breakfast × Social Good Special (Data Village): Why Does My AI Act Like It’s Always Its First Day at Work? (Total 29)2026-11-05 AppWorks talk (Total 30)2026-11-11 Generative AI Conference Meetup 2026 (Total 31)2026-11-15 Taiwan AEC Open-Source Skills Knowledge Base Meetup #4: AI Agent Memory and Knowledge Applications (Total 32)2026-12-18 Talk for a startup development association (Total 33)
2026.05.24

WiDS Taiwan (Women in Data Science)

Five AIs on My Team: Building Your Agent Team

Recap →
2026.06.04

COMPUTEX / InnoVEX 2026: Open Source Team Taiwan Pavilion

2026.07.16

AWS Summit Taipei 2026

From AI Sessions to Shared Memory: Knowledge Infrastructure for AI-Native Teams

Recap →
2026.07.16

Agile Taiwan

From OpenClaw to Cairn: When Agents Join the Team, How Should Memory Be Shared?

Recap →
2026.08.18

Taiwan AI Annual Conference 2026

From AI Sessions to Team Memory: Designing and Building Cairn

2026.12.18 · Upcoming

Talk for a startup development association

Details soon

Topics and how to invite me →

Base camps · Work

Things I’ve built

Two products, plus open-source projects that usually come with a tutorial.

Base camp · In development

Cairn

A cairn is a pile of stones hikers leave to mark the way for whoever comes next. This map borrows the idea behind Cairn: what each person leaves behind becomes a path someone else can find.

Shared memory for people and AI agents. People leave real experience as field notes, and Moss, a custodian agent, organizes, links and maintains them. Built solo in six days for Anthropic’s global hackathon.

2.4%hackathon acceptance
6 daysbuilt solo
Mosscustodian agent
Watch the demo →
Anthropic hackathon key visualBuild log →
AppWorks Accelerator #31Live

AILogora

A Chinese-first community for AI practitioners, organized around viewpoints and topics: perspective cards, discussions, a knowledge map, and a weekly digest for nearly 4,000 subscribers.

Visit AILogora →
NVIDIA RIVA · LangChain

LLMAvatarTalk

A real-time voice assistant with an animated avatar, built on NVIDIA Riva ASR/TTS, LangChain and Audio2Face. Top 100 in the NVIDIA × LangChain Generative AI Agents Developer Contest.

GitHub →
GraphRAG · Python

GraphRAG Visualization Tutorial

A hands-on guide to Microsoft GraphRAG with visualizations and a prebuilt index, so you can explore it without paying for indexing.

GitHub →
JavaScript · SQLite

cairn-memory

The open-source memory layer behind Cairn: cross-session memory for AI agents where every memory carries a receipt, the exact source text it came from. Inspect it, correct it, or forget it for good.

GitHub →
LangChain · Streamlit

RAG_LangChain_streamlit

A from-scratch retrieval-augmented generation example with Streamlit, LangChain and Llama 2.

GitHub →
Research

LLM Spec Comparison

A side-by-side comparison of mainstream LLMs: ChatGPT, Gemini, Claude, Mistral, Llama and more.

GitHub →
OpenClaw · Markdown

skill-openclaw-map

A map of OpenClaw skills and workflows, gathered into one place you can actually search.

GitHub →

All work →

Press

On OpenClaw and AI agents, in the press

Interviews about OpenClaw, an open-source AI agent, and what it takes to use agents for real.

ChiChieh Huang ▲ SURVEYOR
About · The surveyor

I’m ChiChieh Huang, a founder and AI engineer in Taipei.

I’m building Cairn, shared memory for people and AI agents. Before that I built AILogora, a Chinese-first community for AI practitioners that went through AppWorks Accelerator #31. Since late 2023 I’ve written 107 essays explaining generative AI in Chinese, and in 2026 I gave 32 talks across Taiwan.

Focus
LLM applications, RAG, AI agents, agent memory, speech, 2D/3D avatars
Work with me
Talks, workshops, corporate training, advisory
Contact
[email protected]
More
Full bio GitHub LinkedIn RSS