FROM THE TRENCHES Issue №313

Practical decisions for AI and software

Technical guides to choose models, verify software and plan delivery.

Latest articles

New comparisons, practical guides and engineering lessons.

TrueForge agent harness connecting models, MCP tools, subagents, approvals, sandboxed code and traces AI & Agents

TrueForge Review: Is the Open-Source Agent Harness Production-Ready?

A buyer-focused review of TrueForge, its agent loop, context controls, benchmark economics, deployment modes and enterprise production gaps.

Kevin Riedl
Read
One page served as HTML, as a Markdown mirror and as an llms.txt entry, with a build gate checking all three AI & Agents

Agent-Readable Websites: llms.txt, Markdown Mirrors and What Breaks

What public HTML, crawler controls and optional machine formats actually do, with the conversion failures our build gate catches.

Kevin Riedl
Read
Delivery team size compared across a solo operator, a micro agency and a mid-size agency Leadership & Teams

Micro Agency vs Mid-Size Agency: How Big Should Your Software Partner Be in 2026?

The one-person agency versus micro agency argument is a debate between sellers. Here is the buyer’s version, with the team-size research, the 2026 agency margin data, and the four clauses that make a small partner safe to hire.

Kevin Riedl
Read
Self-hosted screen agent operating a desktop interface inside a single-GPU deployment AI & Agents

Qwen3.8-27B: Self-Hosted Computer-Use Agents Without Exporting Screenshots

An open-weight model now leads desktop-operation benchmarks. What that changes for back-office automation you cannot send to a hosted API.

Kevin Riedl
Read
One English slug per document across four language prefixes, with hreflang links pointing between the versions AI & Agents

Localized URL Slugs vs hreflang: Why We Keep One English Slug

Localized slugs are valid, but they add a mapping layer around hreflang. Why we keep one language-independent slug across four languages, and what that costs us.

Kevin Riedl
Read
An Odoo database beside a team's own integration service, with the call ceiling, the transaction boundary and the upgrade clock drawn between them Delivery & QA

Odoo API integration: the five limits that decide your architecture

Odoo Cloud treats one call per second as acceptable use, every JSON-2 call is its own transaction, and XML-RPC has a removal date. What that forces on an integration.

Kevin Riedl
Read
An agent reading a page on one side and completing a task against an authorized tool surface on the other AI & Agents

Can an AI Agent Use Your Product, or Only Read About It?

Readable means an assistant can cite you. Usable means it can complete a task and be refused when it should be. The second is an authorization problem, not a model problem.

Kevin Riedl
Read
A coding agent repo map split into a committed team artifact and a regenerable local cache AI & Agents

Graft Review 2026: Do Agent Repo Maps Belong in Git?

Graft writes your codebase into linked markdown so agents stop re-exploring it. We checked whether that map really travels through git, what its benchmarks prove and what belongs in version control instead.

Kevin Riedl
Read
A prompt passing through a pseudonymization layer, with the re-identification key staying on the controller side Business & Regulation

LLM Pseudonymization Gateways: Does the Prompt Leave GDPR Scope?

Platforms in this category promise that pseudonymizing a prompt makes a hosted model safe to use. The 2025 CJEU ruling and the EDPB guidance say something narrower. Here is the evidence a buyer actually has to collect.

Kevin Riedl
Read
Inbox, without the noise

Get the next field note

One concise email when we publish. No tracking pixels, and no inbox filler.