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.

165

Articles

Likes

383

Views

Comments

Latest from Data Bricklaying Diary

100 recent articles max

Data Bricklaying Diary

Sep 20, 2026 · Artificial Intelligence

From AI Coding to Business Agents: Transferring Engineering Practices with Business Context

The article explains how R&D teams can transfer AI programming practices — task templates, controlled tools, validation cases, and handover flows — to build business agents, but must first add missing business context: object mapping, rule applicability, and output purpose, illustrated through an after-sales ticket summarization case study.

AI EngineeringAI agentsR&D Practices

0 likes · 17 min read

From AI Coding to Business Agents: Transferring Engineering Practices with Business Context

Data Bricklaying Diary

Sep 20, 2026 · Backend Development

Semantic Decoupling: What Ontology Changes Can (and Can't) Spare You From Rewriting

This article analyzes the practical limits of semantic decoupling through ontology models, showing that while field renames and stable contracts can be isolated via mapping layers, rule changes require verified data and computation, and new actions demand real execution paths — value lies in precisely scoping change impact, not promising zero development.

Anti-Corruption LayerDomain-Driven Designbusiness modeling

0 likes · 24 min read

Semantic Decoupling: What Ontology Changes Can (and Can't) Spare You From Rewriting

Data Bricklaying Diary

Sep 19, 2026 · Artificial Intelligence

Beyond Ontology: The Four Knowledge Types Enterprise Agents Actually Need

This article argues that enterprise AI agents require four distinct knowledge categories—world knowledge, domain semantics, task processes, and real-time situational state—each with separate governance and update cycles, rather than stuffing everything into a single ontology, and illustrates how they converge during task execution using a contract risk detection example.

Agent RuntimeDomain SemanticsKnowledge Architecture

0 likes · 18 min read

Beyond Ontology: The Four Knowledge Types Enterprise Agents Actually Need

Data Bricklaying Diary

Sep 18, 2026 · Artificial Intelligence

Self-Built AI Agents: Which Capabilities Are Worth the Engineering Investment?

The article argues that while building custom AI agent tooling is feasible, teams should first use existing solutions for representative tasks, measure full-lifecycle costs including human intervention and maintenance, and only invest in proprietary development for business-specific knowledge, internal system adapters, and critical control gaps that generic tools cannot address.

AI agentsBusiness LogicDeepSeek Harness

0 likes · 13 min read

Self-Built AI Agents: Which Capabilities Are Worth the Engineering Investment?

Data Bricklaying Diary

Sep 18, 2026 · Artificial Intelligence

High-Quality Data Annotation: Business Rules, Expert Disagreements & Arbitration Mechanisms

This article explains how high-quality data annotation requires defining annotation units, business rules, evidence requirements, and rejection criteria, then using double-blind annotation, disagreement classification, expert arbitration, and version governance to ensure correctness, rather than relying solely on inter-annotator agreement metrics.

Data Annotationactive learningbusiness rules

0 likes · 29 min read

High-Quality Data Annotation: Business Rules, Expert Disagreements & Arbitration Mechanisms

Data Bricklaying Diary

Sep 17, 2026 · R&D Management

Closing the Loop: AI-Native SDLC Feeds Production Reality Back to R&D

This article explains how AI-native SDLC transforms production incidents into continuous development feedback by using deterministic detection for anomalies, AI-assisted diagnosis for evidence gathering, risk-tiered automation paths, protected regression tests from failures, and multi-layer asset updates — all while keeping humans accountable for decisions and releases.

AI-Native SDLCAI-assisted diagnosisIncident Management

0 likes · 12 min read

Closing the Loop: AI-Native SDLC Feeds Production Reality Back to R&D

Data Bricklaying Diary

Sep 17, 2026 · Artificial Intelligence

Why Enterprises Need Business Ontologies Despite LLMs' World Knowledge

This article explains why large language models' general world knowledge cannot replace enterprise business ontologies, which provide versioned, traceable semantic models for object identity, institutional definitions, state validity, and responsibility boundaries within a specific organization.

Business OntologyEnterprise Knowledge ManagementLarge Language Models

0 likes · 19 min read

Why Enterprises Need Business Ontologies Despite LLMs' World Knowledge

Data Bricklaying Diary

Sep 16, 2026 · R&D Management

AI Agents Can Prepare Releases But Must Not Cross Production Gates

This article argues that while AI agents can accelerate code review, release preparation, and controlled deployment steps, they must never autonomously obtain production authorization or accept production risk; instead, different environments require tiered autonomy, production permissions must be short-lived, minimal, and auditable, pre-approved runbooks need strict scope and abort conditions, and version rollback does not equal business recovery.

AI agentsDevOpsProduction Deployment

0 likes · 12 min read

AI Agents Can Prepare Releases But Must Not Cross Production Gates

Data Bricklaying Diary

Sep 16, 2026 · R&D Management

Beyond Architecture Diagrams: FDE's Trade-offs for Boundaries, Integration & Risk

This article explains how Frontline Deployment Engineering (FDE) moves beyond architecture diagrams to make concrete trade-offs on system boundaries, capability placement, integration patterns, risk acceptance, and decision invalidation — using a synthetic manufacturing case to show how authority, contracts, and failure paths must be explicitly owned.

ADRFDEIntegration Patterns

0 likes · 26 min read

Beyond Architecture Diagrams: FDE's Trade-offs for Boundaries, Integration & Risk

Data Bricklaying Diary

Sep 15, 2026 · Backend Development

Beyond 'Everything Is a Plugin': Defining Production Responsibility in Agent Harness

The article argues that a composable agent runtime alone is insufficient for production; it requires action contracts for external side effects, version binding with impact analysis, and independent audit evidence for task completion, with clear responsibility boundaries across runtime, plugins, action gateways, and human operators.

Agent Runtimeaction contractsaudit evidence

0 likes · 18 min read

Beyond 'Everything Is a Plugin': Defining Production Responsibility in Agent Harness