Columns
Engineering & AI Judgment
From a problem-framing method to measurement protocols, AI engineering boundaries, org redesign, and agent collaboration — a research thread on technical and organizational judgment.
- Chapter 1 · Deep research · 19 min read
The Five Lenses: Connection-Oriented Problem Location
A framework paper that advances structured thinking from the generic expression of "conclusion first, mutually exclusive and collectively exhaustive" into a four-stage problem-location method — Structure → Surface connections → Infer root problem → Find solution — with the causal, duality, dialectical, position, and interest lenses at its core, offering a testable training protocol and its boundaries of use.
- Chapter 2 · Deep research · 11 min read
Data Measurement as Organizational Protocol: Definitions, Measurement, Tiering, and Retrospectives
A research-design and protocol paper: reframing data measurement from what belongs on the dashboard into a recomputable, reviewable, decision-supporting collaboration protocol, with minimal mechanisms for definitions, measurement units, metric tiering, dictionaries, and retrospectives.
- Chapter 3 · Deep research · 9 min read
Developer Productivity Is Not a Tool Catalog, but a Feedback System
A research synthesis and position paper: defining engineering productivity as a continuously shortening "propose change — get trustworthy feedback — correct safely" loop, with metrics, a default path, and a falsifiable pilot protocol.
- Chapter 4 · Deep research · 9 min read
Defining the Boundaries Before Bringing AI Capability into an Engineering Organization
A position paper based on the author's experience driving AI infrastructure and pilots in a real engineering organization: it proposes four kinds of boundaries—context, responsibility, authorization, and measurement—and offers a falsifiable pilot protocol plus a minimal harness as a demonstration of putting them into practice.
- Chapter 5 · Deep research · 14 min read
When AI Lowers Workflow Barriers: How to Redivide Functional Lines and Business Lines
A position paper: functional lines were efficient in an era of highly specialized work; once tools lower skill barriers, organizations should be redesigned around end-to-end business capability. It provides the mechanism, a spectrum of organizational forms, judgment signals, the two Chinese and American starting points, talent needs, and a falsifiable pilot protocol.
- Chapter 6 · Deep research · 13 min read
Decompose First, Then Schedule: A Review of Multi-Model Task Decomposition, Capability Switching, and Subtask Routing
This paper examines task decomposition, subtask capability classification, model selection, and bounded execution-time fallback, while reducing the broader Agent scheduling problem to an engineering slice of dsh-quota-router.
- Chapter 7 · Deep research · 13 min read
Making an Agent a Collaborable Object: State, Feedback, and Result Confirmation
Based on ChatLab design work, this paper records how presence, response, waiting, and delivery patterns from human collaboration can make Agent runtime states easier to understand; it is an HCI design observation about feedback, evidence, intervention, and result confirmation.
- Chapter 8 · Deep research · 13 min read
Let the Agent Execute: Emotional Adaptation, Trust Calibration, and Human Relief from Execution Pressure
Based on long-task use, this paper examines how an Agent can adapt state, evidence, and action feedback to uncertainty, waiting, failure, and takeover needs, while documenting how delegated execution changes human pressure, trust, and capability boundaries.
- Chapter 9 · Deep research · 29 min read
The Laws of Human Motivation: The Situational Motivation Model
A theoretical synthesis, conceptual model, and application framework: it redefines motivation from "whether someone has drive" into a situated process of "whether motivation is generated, whether it can be converted into behavior, and whether it can be sustained," integrating self-determination theory, expectancy-value, goal setting, self-efficacy, and behavior models; bringing in perspectives from developmental psychology, business management, and organizational and occupational psychology to propose a three-layer "Situational Motivation Model" with a "growth axis," which it concretizes in the workplace into six components — task, goal, incentive, feedback, development path, and leadership/relationship — connecting to measurable outcomes such as engagement, commitment, performance, burnout, and turnover, introducing the enterprise life cycle as an enterprise-level moderating variable, and ultimately yielding nine falsifiable laws of motivation.
- Chapter 10 · Short · 5 min read
AI Does Not Automatically Create Productivity: From Local Acceleration to System Value
Starting from a multi-system integration experience, this essay distinguishes creation, task efficiency, organizational productivity, and business value—and asks where value and cost actually come from when AI enters a complex system.