AI builder. Founder. Operator log.
Carlos KiK
RSS FeedI write from inside the build: AI, persistent memory, founder reality, and the future of human-machine connection. I am currently building KAi at Digital Human Corporation in Seoul.
This site is the public notebook for the work around it: what is changing in AI, what I am learning while building, and what it costs to keep going when the obvious path is not available.
Current Bet
DHC, KAi, persistent memory, and the long fight to build useful AI without turning people into engagement inventory.
The working thesis is simple: the next serious AI companion category is not about hotter chatbots. It is about memory, trust, privacy, continuity, and whether the system helps people return to real life with more clarity.
ckik.dev tracks that thesis in public through daily AI analysis, founder notes, and the story behind the company. Some pieces are fast dispatches. Some are long arcs. The center of gravity is always the same: what is real, what matters, and what I would bet my own time on.
For the company-side version, read The Architect .
Follow The Lab
AI Intel
The daily signal: models, agents, robotics, memory, compute, security, and the product moves that matter.
Startup Life
Founder reality from Seoul: public story arcs, technical scars, hard lessons, and the cost of building for real.
DHC And KAi
The company and companion I am building around persistent memory, privacy, and adult AI support.
- Field notes from the AI frontier
- Decision ledgers from the build
- Founder stories with the dangerous parts filtered correctly
- Unsung heroes behind the systems everyone uses
Recent Posts
-
Meta Bought Conductor, and the Parallel Agent Layer Just Became a Buyable Category
A 2024 startup that orchestrated coding agents from Anthropic and OpenAI sold to Meta. The cloud shuts down Nov 20. Neutral tooling died with it.
-
OpenAI's Pipeline Found 948 Ways to Become Root in the Linux Kernel
2,308 candidate findings, 1,724 reproduced, 948 working privilege escalation exploits, $1.2M in API credits. The verification loop is the actual story.
-
OpenAI Bought Pierre and Is Quietly Building Its Own Git Plumbing
Pando pushed a petabyte of Git data a day before OpenAI bought Pierre's Code Storage. Source control is becoming the layer labs fight over.
-
Anthropic's Multi-Agent Arc: Subagents In July 2025 To Agent Teams In 2026
Claude Code subagents shipped July 2025 with separate context windows and tool permissions. The same idea now runs as agent teams and dynamic workflows.
-
Nvidia Open-Sourced Auto Ontology, a Framework for the Definitions Agents Depend On
Auto Ontology on GitHub under Apache 2.0: a Python framework for semantic graph reasoning layers, reconciling conflicting metrics, serving agents via MCP.
-
OpenAI Is Merging Chat and Work Into One Surface, With One Token Pool
A user email on November 2 confirmed Chat and Work tabs merge with a shared token pool. Codex stays separate. The consumer-work split was a fiction.
-
Reality Tested: Dyson CameraJet
Why I bought a S$699 toothbrush after my last S$500 smart toothbrush failed me—and what happened when Dyson's claims met Day 1 reality.
-
Ox Alpha Is a Frontier Model Nobody Will Claim. That Silence Is the Story.
OpenRouter listed an anonymous frontier model called Ox Alpha with a 1M context, video input, and a free week. Four anonymous models before it were all claimed by Chinese labs.