BestHub
Sign in
Data Bricklaying Diary
Author

Data Bricklaying Diary

Records practices, thoughts, and pitfalls on the data grunt-work journey, sharing content on data platforms, data analysis, data processing, data governance, knowledge graphs, and more. Less theory, more hands‑on, making complex data technologies simple.

86
Articles
0
Likes
24
Views
0
Comments
Recent Articles

Latest from Data Bricklaying Diary

86 recent articles
Data Bricklaying Diary
Data Bricklaying Diary
Jul 28, 2026 · Product Management

Don't Ask Users What They Want—Observe Work to Find Real AI Opportunities

This article argues that effective AI opportunity discovery requires shadowing real work to uncover friction points like repetitive information gathering and delayed exception detection, rather than asking users for feature requests, and introduces the AI Opportunity Card framework to evaluate and prioritize pilot projects based on stable evidence, human boundaries, and verifiable outcomes.

AI opportunity cardAI opportunity discoverybusiness research
0 likes · 9 min read
Don't Ask Users What They Want—Observe Work to Find Real AI Opportunities
Data Bricklaying Diary
Data Bricklaying Diary
Jul 27, 2026 · R&D Management

Defining the First AI Pilot: From Value Narrative to Minimum Value Loop

The article argues that the first AI pilot should start with a value narrative defining role, scenario, current state, target state, and verifiable metrics, then select a scenario meeting five feasibility conditions to run a minimum value loop of role, action, AI assistance, human confirmation, and metrics, enabling a go/expand/adjust/stop decision.

AI implementationAI projectR&D management
0 likes · 9 min read
Defining the First AI Pilot: From Value Narrative to Minimum Value Loop
Data Bricklaying Diary
Data Bricklaying Diary
Jul 26, 2026 · Artificial Intelligence

From Harness to Loop: Engineering AI Agents That Converge, Not Just Execute

The article distinguishes Harness Engineering (providing controlled execution environments for AI agents) from Loop Engineering (using verification evidence to classify deviations, adjust plans, and drive tasks to verified completion), detailing required state management, evidence-based stopping conditions, and a concrete rate-limiting example.

AI Coding AgentsHarness EngineeringLoop Engineering
0 likes · 15 min read
From Harness to Loop: Engineering AI Agents That Converge, Not Just Execute
Data Bricklaying Diary
Data Bricklaying Diary
Jul 25, 2026 · Industry Insights

Why Court AI Must Start with Case Reading, Not Document Generation

This article argues that court AI systems should prioritize case file reading, issue identification, and evidence understanding over document generation, proposing a chain of specialized agents built on a shared judicial semantic foundation to embed AI assistance within existing trial workflows while maintaining procedural boundaries and human oversight.

AI agentscourt AIevidence review
0 likes · 18 min read
Why Court AI Must Start with Case Reading, Not Document Generation
Data Bricklaying Diary
Data Bricklaying Diary
Jul 23, 2026 · Artificial Intelligence

Operating High-Quality Datasets as Continuous Data Products for AI

This article presents a six-step framework for operating high-quality datasets as continuous data products, covering responsibility assignment, version baselines, quality and AI effect monitoring, feedback-to-candidate pipelines, controlled release strategies, and retirement mechanisms to ensure datasets evolve with business, models, and risk boundaries.

AI data productsAgentOpsDataOps
0 likes · 14 min read
Operating High-Quality Datasets as Continuous Data Products for AI
Data Bricklaying Diary
Data Bricklaying Diary
Jul 22, 2026 · Artificial Intelligence

Beyond Accuracy: A Five-Layer Framework for Evaluating High-Quality AI Datasets

This article presents a five-layer evaluation framework for high-quality AI datasets—covering basic data quality, business semantics, task adaptation, AI application effects, and trustworthy operations—emphasizing task-specific validation over generic metrics and advocating admission vs. optimization metrics with automated, expert, and task-based verification.

AI data qualityRAG evaluationadmission metrics
0 likes · 13 min read
Beyond Accuracy: A Five-Layer Framework for Evaluating High-Quality AI Datasets
Data Bricklaying Diary
Data Bricklaying Diary
Jul 20, 2026 · Big Data

High-Quality Datasets: Beyond Cleaned Data for AI Tasks

This article defines high-quality datasets as task-oriented data products with business semantics, evidence traceability, usage boundaries, and continuous governance — not merely cleaned data — and explains why they are essential for reliable AI training, RAG, and Agent applications.

AI data preparationAgentData Governance
0 likes · 13 min read
High-Quality Datasets: Beyond Cleaned Data for AI Tasks
Data Bricklaying Diary
Data Bricklaying Diary
Jul 19, 2026 · R&D Management

AI Programming Era: Design Judgment Is the Real Scarcity, Not Code

The article argues that AI lowers code generation cost but amplifies the need for clear requirements, architecture, and verification; it advocates Spec-Driven Development (SDD) with controlled agents and test feedback loops, and shares a practical workflow separating design, implementation, and testing roles to ensure systems are correctly designed, implemented, and validated.

AI programmingHarness EngineeringLoop Engineering
0 likes · 20 min read
AI Programming Era: Design Judgment Is the Real Scarcity, Not Code