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
Sep 2, 2026 · Backend Development

Independent Deployment ≠ Clean Internals: Isolating Business Rules from Technical Details

This article explains why independently deployed services can still have chaotic internal architecture, and demonstrates how to separate business rules, use cases, ports, and adapters to ensure dependencies point inward, enabling testable, evolvable systems without mechanically applying layered templates.

business rules isolationclean architecturedependency inversion
0 likes · 22 min read
Independent Deployment ≠ Clean Internals: Isolating Business Rules from Technical Details
Data Bricklaying Diary
Data Bricklaying Diary
Sep 2, 2026 · Artificial Intelligence

Why Random Splits Fail for Enterprise Data: Preventing Time, Entity & Business Leakage

This article explains why random row-based splitting creates false generalization in enterprise ML, detailing how to define prediction timepoints, combine business entity grouping with time windows, and implement five-layer leakage checks to ensure test data truly simulates unseen production data.

ML evaluationbusiness groupingdata leakage
0 likes · 23 min read
Why Random Splits Fail for Enterprise Data: Preventing Time, Entity & Business Leakage
Data Bricklaying Diary
Data Bricklaying Diary
Sep 1, 2026 · Artificial Intelligence

Dependency Injection ≠ Dynamic Replacement: Managing Agent Capability Recomposition at Runtime

This article explains why dependency injection only handles startup binding, and how a composable agent runtime must detect dependency changes, freeze affected consumers, validate compatibility, rebind with policies, invalidate stale evidence, and escalate to humans—ensuring capabilities never operate on outdated assumptions.

Agent RuntimeComponent Dependency GraphDependency Injection
0 likes · 18 min read
Dependency Injection ≠ Dynamic Replacement: Managing Agent Capability Recomposition at Runtime
Data Bricklaying Diary
Data Bricklaying Diary
Sep 1, 2026 · Fundamentals

Beyond Consistency Checks: A 7-Layer Acceptance Framework for Ontology Models

This article presents a seven-layer acceptance framework for ontology models that moves beyond consistency reasoning and expert review, covering capability questions, semantic constraints, real data mapping, rule execution with unknown states, exception handling and action contracts, version traceability, and controlled business outcomes with multi-role sign-off.

business-outcomesconsistency-reasoningexpert-review
0 likes · 26 min read
Beyond Consistency Checks: A 7-Layer Acceptance Framework for Ontology Models
Data Bricklaying Diary
Data Bricklaying Diary
Aug 30, 2026 · Industry Insights

Palantir Ontology Isn't Just Schema Design — It's a Runtime Semantic Platform

The article argues Palantir Ontology's primitives aren't novel, but its value comes from organizing them into an engineering system that connects business semantics, real data, controlled actions, permissions, and feedback loops, enabling reusable, governable capabilities across queries, applications, and agents — not merely a static model diagram.

PalantirPlatform EngineeringRuntime System
0 likes · 19 min read
Palantir Ontology Isn't Just Schema Design — It's a Runtime Semantic Platform
Data Bricklaying Diary
Data Bricklaying Diary
Aug 29, 2026 · Operations

From LLMOps to AgentOps: Operating Enterprise Agents Across Full Task Lifecycles

This article argues that enterprises need AgentOps, not just LLMOps, to manage AI agents that execute multi-step tasks with tools, state, and human oversight, detailing six key capabilities: task identity, state checkpoints, component versioning, end-to-end observability, task-level evaluation, and human-in-the-loop as a first-class operational state.

AI agentsAgentOpsLLMOps
0 likes · 15 min read
From LLMOps to AgentOps: Operating Enterprise Agents Across Full Task Lifecycles
Data Bricklaying Diary
Data Bricklaying Diary
Aug 28, 2026 · Backend Development

Why Service Boundaries Fail: Aligning Business Rules, Data Ownership & Team Topology

This article explains why effective service boundaries require aligning business rules, data ownership, change rhythms, and team responsibilities—not just drawing boxes around nouns—using a refund processing case study to compare modular monolith vs. microservice trade-offs and verify boundaries through concrete evidence.

architecture-designbounded-contextdata-ownership
0 likes · 21 min read
Why Service Boundaries Fail: Aligning Business Rules, Data Ownership & Team Topology
Data Bricklaying Diary
Data Bricklaying Diary
Aug 28, 2026 · Industry Insights

FDE Isn't On-Site Development: Why Enterprise AI Needs Ownership from Problem to Production

This article defines the Forward Deployed Engineer (FDE) as a closed-loop responsibility model for enterprise AI, distinguishing it from on-site development by emphasizing end-to-end accountability from ambiguous problems through production validation, handover, and product feedback, and outlines when and why organizations need this role.

AI DeploymentFDEForward Deployed Engineer
0 likes · 27 min read
FDE Isn't On-Site Development: Why Enterprise AI Needs Ownership from Problem to Production
Data Bricklaying Diary
Data Bricklaying Diary
Aug 27, 2026 · R&D Management

System Architecture Isn't Tech Selection: The Critical Questions for Production Readiness

This article argues that system architecture starts from business goals and quality attributes, not technology choices, using a refund processing example to illustrate trade-offs across system boundaries, collaboration patterns, and six architectural perspectives, emphasizing that observability, fault tolerance, and security must be designed in and validated with independent evidence throughout the system's lifecycle.

architectural trade-offsevidence-based validationmicroservices
0 likes · 20 min read
System Architecture Isn't Tech Selection: The Critical Questions for Production Readiness