Professional architect sharing high‑quality architecture insights. Topics include high‑availability, high‑performance, high‑stability architectures, big data, machine learning, Java, system and distributed architecture, AI, and practical large‑scale architecture case studies. Open to ideas‑driven architects who enjoy sharing and learning.
The article explains how an ontology—defining objects, relationships, constraints, and inferable boundaries within a domain—serves as a semantic control plane between the fact and action layers of an agent‑driven system, ensuring consistent interpretation, validation, and lifecycle management of business facts.
The article analyzes TencentDB Agent Memory’s design, breaking down its three‑stage write‑read‑governance pipeline, four‑level L0‑L3 hierarchy, object types, recall strategies, conflict handling, sharing rules, long‑task traceability, and practical testing guidelines to ensure past information helps future decisions while preserving provenance and error correction.
WorkBuddy Bench reveals why fixing a bug does not equal finishing a task, proposing a four‑layer completion model and a reproducible benchmark that evaluates agents across Code, Web, Office, and Security workspaces, showing how prompts, context, harnesses, loops and graphs must be verified to claim true completion.
As LLM agents gain the ability to read code, modify files, send emails and call APIs, ArbiterOS introduces a runtime governance layer that turns model intents into structured, traceable instructions, enabling policies to approve, block, or request confirmation before any high‑risk action is executed.
The article walks through OpenWorker’s four‑layer runtime architecture, detailing how a desktop UI, a Python agent service, a permission engine, and resource/execution layers cooperate to turn a single user request into a visible, auditable, and openable deliverable, while comparing it to Tencent’s WorkBuddy.
The article breaks down the six essential components of Claude Code and Codex—real‑time repository context, prompt caching, tools & permissions, context governance, session memory, and sub‑agents—showing how they align task state, action boundaries, and evidence to make coding agents reliable in real development workflows.
The article breaks down agent engineering into five concentric layers—Prompt, Context, Harness, Loop, and Graph—explaining how each governs task definition, current facts, execution environment, iterative exploration, and stable cross‑task relationships, illustrated with a payment‑callback fix case and AutoResearch insights.
When AI‑assisted coding is introduced, teams often struggle to preserve the hidden engineering expertise of top performers, so the article proposes turning that tacit know‑how into reusable, inspectable, and trim‑able Agent processes that can be documented, run, and continuously improved.
The article analyses Claude Code’s four loop types—turn‑based, goal‑based, time‑based and proactive—as a four‑level control‑delegation ladder, explains how each level shifts responsibility from human to system, and offers a gradual adoption path with timely downgrade strategies to keep automation reliable.
Anthropic’s Claude Code team reports an eight‑fold increase in engineer code output, but the article explains that the real shift is from writing code to verification, with specs stored in repos, routines that merge feedback, PRs and metrics, a bad/sad quality taxonomy, six concrete interfaces, and new challenges around context switching, accountability and role boundaries.
