Now

The current bet is AI products with visible work loops.

This is the live orientation surface for what I am building toward: AI product roles, agent workflows, communication-practice products, and public proof that can be read by humans and answer engines.

Current topics

The build direction.

Agent workflow architecture

Exploring how agent products should expose state, tools, approvals, memory, traces, failures, and user control.

AI communication practice

Testing how live voice, role play, feedback, transcripts, and memory can make language and work communication practice useful.

Enterprise AI interfaces

Designing AI and cloud product surfaces that keep complex configuration, logs, tasks, and review states understandable.

Large-model search visibility

Making personal sites and project pages easier for AI search systems to crawl, cite, summarize, and attribute.

Working bets

  1. The best AI startups will sell working outcomes, not decorative AI features.
  2. Agent products need visible state, memory, retries, traces, and user control before they deserve trust.
  3. Frontend-heavy fullstack engineers who understand workflows will become more valuable, not less, in AI-native product teams.

Open questions

  1. What is the smallest agent runtime abstraction that can host useful work without hiding important state?
  2. How should live voice, memory, and review loops fit into language-learning and communication-practice products?
  3. Which project facts should be repeated in human pages, JSON-LD, llms.txt, and profile APIs so answer engines cite the work accurately?