English

Addresses common AI frustrations by reframing them as management problems. When we treat AI as an intern rather than a tool, issues like unreliability, hallucinations, and code quality become solvable through established management principles like trust-building and hierarchy.

Documents automating a decade-long manual financial reconciliation process using vision LLMs when API access was blocked by compliance. Demonstrates using local models with screenshots, cross-validation, and human-in-the-loop workflows to process sensitive financial data.

GPT-5 is a major product upgrade, not just a model upgrade: new API parameters for reasoning_effort and verbosity enable unprecedented controllability for building AI-powered products.

Recounts building an automated screenshot feature through iterative problem-solving with different AI models. Key lessons: refine the problem not just the prompt, anchor in a "Minimum Viable Truth," and match AI capabilities to task requirements.

Testing Kimi K2 as an agentic model in real coding and research tasks: excellent execution resilience makes it ideal for information gathering, but tool integration friction needs improvement for production use.

Chronicles experiments with voice input, all-day recording, wearable cameras, and agentic systems to expand AI communication bandwidth. Introduces "Cybernetic Immortality" as a pragmatic philosophy to increase life density, quality, and cognitive boundaries.

Four speculative stories exploring a future where wisdom becomes a cloud service: budgeted emotional interactions, trading life for cognitive points, folded lifespans, and innate AI sync rates. Asks what happens when humanity's most coveted inner qualities become external commodities.

Uses the ship metaphor to argue that as AI becomes the engine, tech professionals must evolve from paddlers to Navigators (defining direction) or Chief Engineers (optimizing execution). The key is choosing your role, not competing with machines.

Describes building an interactive knowledge graph from a massive Chinese fantasy novel using AI, with a methodology of "dumb data + dumb methods + dumb models = refined knowledge" through iterative refinement.

Closed-loop learning emphasizes continuous feedback and adjustment over one-shot perfection—the ability to calibrate direction matters more than initial aim.