FROM THE TRENCHES Issue №288

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.

DeepSeek V4 Flash 0731 model weights fitted into GB10, Strix Halo and Gorgon Halo unified memory AI & Agents

DeepSeek V4 Flash 0731 on One AI PC: What Actually Works?

Can one GB10, Strix Halo or Gorgon Halo run DeepSeek V4 Flash 0731? A buyer-focused reality check on memory, quantization, privacy, observability and production readiness.

Kevin Riedl
Read
QM gives employees and shared Slack rooms isolated AI agent workspaces connected to one policy-controlled core AI & Agents

YC QM Agent Review: Is Quartermaster Ready for Work?

Quartermaster gives employees and shared rooms isolated AI agent workspaces. Review QM's GitHub architecture, security postures, deployment ownership and pilot fit.

Kevin Riedl
Read
jcode and Claude Code agent harnesses compared across memory, startup, context and team adoption AI & Agents

jcode vs Claude Code: Is the Rust Harness Worth Switching To?

The viral RAM numbers are real but easy to misread. A fact-checked buyer review of jcode's Rust architecture, passive memory, agent-grep, self-dev mode and team adoption risks.

Kevin Riedl
Read
Lightpanda headless browser production evaluation for AI agents AI & Agents

Lightpanda Browser for AI Agents: Production Guide

Lightpanda reports 9x faster execution and 16x lower memory than Chrome. See what the benchmark proves, where compatibility ends, and how to run a safe production pilot.

Kevin Riedl
Read
Graph engineering connects AI agent workflows, experiment lineage and shared knowledge AI & Agents

Graph Engineering for AI Agents: When Does a Knowledge Graph Pay Off?

A buyer-focused guide to execution graphs, experiment DAGs and knowledge graphs, including when to skip them, what provenance requires and how to run a measured pilot.

Kevin Riedl
Read
Multi-model AI coding agent stack with one orchestrator, specialist workers and verification gates AI & Agents

Multi-Model AI Coding Agent Stack: A Team Buying Guide

Separate the harness from the models, route work by role and measure cost per accepted change. A vendor-neutral architecture, ROI scorecard and 30-day rollout plan.

Kevin Riedl
Read
YC application technical readiness checklist with demo, metrics and founder ownership checks Product & MVP

YC Application Technical Readiness: What to Build Before You Apply

YC does not require an MVP, but a working demo, real user evidence and founder-owned technical progress can sharpen your application. Use this checklist to build proof instead of polish.

Kevin Riedl
Read
A stateful MCP session architecture changes into stateless requests that can reach any server instance behind a load balancer AI & Agents

Is MCP Stateless Now? Your Server Migration Checklist

MCP 2026-07-28 removes protocol sessions and the initialize handshake. See what stateless really means, what still keeps state and what to change in your own MCP server.

Kevin Riedl
Read
An AI agent reaches customer data through MCP and an API, while policy checks at the record and field level form the real security boundary AI & Agents

MCP Is Not a Security Boundary: Protect Agent Data

MCP can narrow an agent's tool interface, but it cannot protect a customer record the agent can reach another way. A practical model for identity, egress, object-level policy, field protection and audit.

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.