# Daniel Kim > Personal site and blog by Daniel Kim — AI for knowledge work, media tech (OTT/DRM/streaming), > and building in public. Bilingual EN/KO. This index covers the English content. ## Static pages - [Home](https://danielkimdev.com/index.md): Overview and featured work - [About](https://danielkimdev.com/about.md): Background, career, and what I work on - [Blog](https://danielkimdev.com/blog/index.md): All blog posts - [Portfolio](https://danielkimdev.com/portfolio/index.md): Work portfolio ## Blog posts - [Why I Added an AI Memory Tool and Then Removed It](https://danielkimdev.com/blog/ai-memory-tool-removed.md): I connected Mem0 to manage context across AI coding agent sessions, then removed it. Here's why two plain markdown files turned out to be enough. - [The Myth That You Have to Read Everything: Weekly Review and a Slack Safety Net](https://danielkimdev.com/blog/building-llm-pkm-in-public-ep10.md): Of 128 wiki pages, I had personally read almost none. When AI builds every connection, the knowledge graph belongs to the AI — it just lives under your name. How I built a weekly review workflow and a Slack safety net to keep my own thinking in the loop. - [What a Five-Month Stall Taught Me About Working With AI](https://danielkimdev.com/blog/whatifclassics-fable5-revival.md): Returning to What If Classics after five months with Fable clarified something about AI-assisted work: the bottleneck isn't execution — it's knowing how to package a problem before you hand it over. - [Five Sources In, I Was Rubber-Stamping: Breaking the Browse-and-Accept Pattern](https://danielkimdev.com/blog/building-llm-pkm-in-public-ep9.md): One source at a time, I reviewed carefully. Seven at once, and by the fifth I was scanning titles and rubber-stamping. How I recognized the browse-and-accept failure mode — and why splitting ingest into three separate stages fixed it. - [Ten of Twelve Were Guesses: Auditing the Seeds of a Deterministic Registry](https://danielkimdev.com/blog/building-llm-pkm-in-public-ep8.md): Twelve entries seeded the source registry. Two had real evidence behind them. Ten were AI guesses stored in a file that looked like confirmed facts. How I reconstructed evidence from the ingest history — and what it means when confirmed and unconfirmed data look identical in a system you built to prevent guessing. - [Why This Digital Garden Exists](https://danielkimdev.com/blog/welcome-digital-garden.md): A public notebook on AI for knowledge work, grown by someone who spent two decades shipping and documenting hard systems. - [Agent Readiness: Writing Docs Machines Can Actually Use](https://danielkimdev.com/blog/agent-readiness.md): Technical documentation has a second audience now: AI agents. What agent readiness means, where docs fail agents, and how to measure it. - [The Fix I Trusted Had the Same Flaw: Giving Source Judgment a Real Memory](https://danielkimdev.com/blog/building-llm-pkm-in-public-ep7.md): Source trust judgments kept drifting between sessions, so I built a registry to make them persistent. The two-axis framework behind that design, the author-level extension that strengthened it — and the uncomfortable thing I found in the seed data I'd used to populate it. - [I Thought It Was My Thinking: How AI Quietly Takes Over Your Judgment](https://danielkimdev.com/blog/building-llm-pkm-in-public-ep6.md): Every summary Claude returned seemed obvious — until I realized my own impressions were being shaped by Claude's framing before I'd formed any of my own. What Sofia Quintero's essay helped me see, and how I built a friction step into my ingest workflow. - [One Space, Three Roles: When Different Audiences Need Different Rules](https://danielkimdev.com/blog/building-llm-pkm-in-public-ep5.md): Personal notes, reusable templates, and public writing lived in the same space — and an AI agent had no reliable way to know which rules applied where. Why I split the vault into four explicit layers, and what the separation made visible. - [More Plugins, More Power? The Hidden Cost of a Bloated AI Config](https://danielkimdev.com/blog/building-llm-pkm-in-public-ep4.md): Opening a wiki session, I saw 50,000 tokens already consumed — before touching a single file. A 93 KB development framework had been loading into every session regardless of the task. What I learned about keeping AI context lean. - [What Insight Synthesis Actually Looks Like: The One Point Where DRM Expertise Meets the AI Agent Economy](https://danielkimdev.com/blog/building-llm-pkm-in-public-ep3.md): Looking for where DRM expertise actually meets the AI agent economy, my first answer was wrong. Here's the retraction, and the one real connection point regulation is already making mandatory. - [From Data Hoarder to Knowledge Architect: Bad Sources Make a Bad Wiki](https://danielkimdev.com/blog/building-llm-pkm-in-public-ep2.md): I used to think feeding articles to AI and letting it fill my wiki counted as building knowledge. Here's how bad sources turned that into a well-organized trash can, and the three-part quality check I built to stop it. - [I Started Organizing a Decade of Knowledge with AI — Why and Where to Begin](https://danielkimdev.com/blog/building-llm-pkm-in-public-ep1.md): Starting a public series on building a personal LLM-Wiki out of a decade of OTT and DRM experience. Why AI can summarize but not reflect, and what the system looks like a few weeks in. ## Portfolio - [AI for Knowledge Work & Agent-Ready Docs](https://danielkimdev.com/portfolio/ai-knowledge-work.md): Experiments and writing on AI knowledge management, automation, and making technical documentation AI- and agent-ready. - [This Digital Garden — an LLM-wiki & PKM](https://danielkimdev.com/portfolio/digital-garden.md): The personal-knowledge-management system and AI tooling behind this very site — a Karpathy-style digital garden. - [What If Classics — Interactive Fiction & Generative AI Pipelines](https://danielkimdev.com/portfolio/whatifclassics.md): A "choose your story" library of public-domain classics — branching endings and MBTI character cards — and the text-and-image generation pipeline behind it. - [Technical Writing, Developer Relations & Product](https://danielkimdev.com/portfolio/talks-writing-devrel.md): Developer-facing writing, documentation, and conference talks in media-tech — plus Product Owner for DoveRunner's Platform Ops team, driving requirements across DRM, Watermarking, App Security, and the consolidated customer console. - [Content Security — Multi-DRM & Forensic Watermarking](https://danielkimdev.com/portfolio/content-security.md): Owned the multi-DRM and forensic-watermarking content-protection products (PallyCon SaaS) for global OTT and streaming customers. - [Career — 11+ years, developer to product & DevRel](https://danielkimdev.com/portfolio/career.md): 11+ years in C/C++ and Java, then a move from writing code to shaping products and developer relations in media-tech.