This site is also available as markdown for AI agents. For a complete index of all pages, see the site index at /llms.txt

Daniel Kim

Exploring how AI organizes knowledge, automates busywork, and creates documentation machines can read.

~/daniel/focus

$ cat focus.md

  • AI Knowledge Management
  • Automation
  • AI-Ready Technical Docs
DIGITAL GARDEN

What I'm working on

View portfolio →

How this blog gets written

~/daniel/pipeline.sh

$ ./ingest.sh | tee wiki.log

Every post here starts as a private note, not a blank page. This is the pipeline that turns what I read into what I publish.

  1. 01IngestArticles, docs, and notes come in — tagged, checked for duplicates, logged.
  2. 02Private WikiSources build a cross-linked knowledge base that keeps growing over time.
  3. 03Story ScoutingA synthesis, milestone, or hard-won lesson gets flagged as a post idea.
  4. 04Draft → Review → PublishWritten in my voice, passed through editorial review, then published here.

Everything upstream stays private. Only finished, reviewed writing becomes a post.

Latest writing

7 min read

Adding a Newsletter to My Blog on Kit's Free Plan

My blog only had an RSS feed, so I added email subscription with Kit's free plan. Here's why I picked Kit, what caused the confirmation emails to land in spam and how I fixed it, what the free plan blocks, and how a Cloudflare Worker uses the Kit API to draft a broadcast for every new post.

AutomationSolopreneur

7 min read

Adding a Lightweight Local Search to a 2-Million-Word Wiki

My wiki's raw sources passed 2 million words, and reading whole index files started to break down. Here is how I built hybrid search on SQLite and static embeddings after ruling out a heavier tool, and how I got agents to use it without being told the command.

PKMAI & LLMsAI Knowledge Management

All posts →