Topic

Enterprise software architecture, and the complexity you have to earn

Most architecture advice is written for a scale the reader does not have yet. These pieces argue about sequencing: what a schema registry actually buys you, when a service mesh stops being premature, and what identity costs to self-host.

92 pieces

Teardowns

20

China-Available LLM APIs From Mainland Networks: The Real Map Three gates decide what you can call from a mainland connection: the network, the account and the payment rail. Only one is a firewall, and it is the one that matters least.
Keycloak Single Sign-On Across Distributed Sites, and the Parts That Bite Realms, PKCE, token clocks, brokering and the database nobody budgeted for. What actually breaks when single sign-on spans a dozen sites, and what to decide before it does.
Harness Engineering: The AI Agent Guardrails Nobody Demos The model is a component you swap. The harness around it, tool schemas, validation, sandboxes, budgets, permissions and audit logs, is where nearly every production AI incident actually starts.
Tool calling reliability: DeepSeek, Qwen, Kimi open weights and what to check before you build Four separate properties hide inside the phrase "tool calling works". Here is what each of the three Chinese open-weight families actually documents, where the calls break and how to test it yourself.
How to Test LLM Applications Without Calling the Model The model is the only part of an LLM application that cannot be tested deterministically. Everything around it can be, and that surrounding code is where most production incidents actually start.
Agent Step Budget and Stop Conditions: What Actually Holds in Production Unbounded agent loops are financial incidents waiting to be billed. How to set a step ceiling against a task class, detect stagnation and bound cost separately from steps.
What Kimi, DeepSeek, Qwen and Doubao actually cost from Shenzhen Benchmark parity is not workflow parity. The four Chinese frontier models are genuinely competitive on price and quality, and the reasons a Western team still cannot use them are not technical.
Signing Off on AI Agent Security Without Lying to Yourself A language model cannot tell your instructions from the text it retrieves, because both arrive as the same token stream. Every real control follows from accepting that rather than trying to fix it.
Designing APIs for AI Agents When the Caller Cannot Read Your Docs An API read by a model has different requirements from one read by a developer. Names become prompt, errors become instructions and idempotency stops being optional once the retry loop is autonomous.
The agent control loop is where production agents fail Prompt engineering was never the hard part. The loop is: what the agent sees, what it may touch, what stops it and what a step costs. Most agent failures are loop failures, not model failures.
AI Legacy Code Modernization Is Not a Translation Problem Legacy modernization was never mainly a translation problem. The hard parts are comprehension and behaviour preservation, and a model's real value sits almost entirely in the first one.
The AI Data Privacy Checklist I Run Before Shipping a Feature Retention windows, zero-retention gaps, personal data in vector stores and the erasure problem, written for the engineer who has to configure it before Friday.
Six plants, one event log, and what it actually guarantees An event log across distributed plants buys ordering per partition, replayable history and one shared dependency, and most factories would be better served by a database for another five years.
LLM Evals Are the Regression Tests for Probabilistic Software Teams that cannot measure their AI system cannot improve it, so every change becomes an argument. Evals are the regression tests of probabilistic software and the first thing to build.
When to Use a Service Mesh: Day One Is Usually Too Early A mesh gives you enforcement your code cannot. It also charges per pod from day one. What ambient mode changed, what a gateway plus disciplined timeouts still covers and when the tradeoff flips.
Unified Namespace vs Data Historian: Why Neither Replaces the Other A UNS is a real-time topic space. A historian is an archive that discards data on purpose. The retention maths, the Sparkplug detail and the honest architecture where both exist.
How DeepSeek Builds World-Class AI on a Shoestring Budget The real story behind the $5.6 million model that panicked Silicon Valley, explained so you actually understand it.
The Protobuf Schema Registry as the Single Source of Truth The schema file is the contract. Documentation describes it and whichever service was written first is a guess at it. What Buf, field numbers and reserved actually buy you.
What Actually Breaks in Distributed Engineering Teams Three sentences of code review feedback can cost two calendar days when the reviewer is asleep. The pattern of what breaks in distributed engineering work and what actually fixes it.
The break-even on self-hosted open weights is 72 percent utilisation Renting two H100s to serve gpt-oss-120b costs $4,803 a month and beats a serverless endpoint only above 72 percent sustained utilisation, which almost nobody's traffic reaches.

Essays

4

Signals

1

Taalas Never Sold a Chip. AMD Bought the Memory Trick Underneath It The HC1 burns Llama 3.1 8B into silicon and quotes 17,000 tokens a second. Taalas' own product page calls it a technology demonstrator. That phrase is the whole story.

Projects

2

Tools

12 of 65

All 65 matching tools