Loops, graphs & harnesses – getting quality out of a software factory

kept by eddie sep 11

The article explains how to construct a reliable AI‑driven software factory using dual loops, explicit human bottlenecks, and deterministic validation to improve code quality while minimizing manual agent output review.

Company as Code
The author proposes defining a company's structure, roles, and processes as declarative code, enabling versioning, testing, and automation.
also on Terraform, GitHub, #automation, #software-engineering, #developer-tools
What are your programming "hunches" you haven't yet investigated?
Lobsters commenters share uninvestigated programming hunches, including accessibility-first GUI toolkits, 32-bit pointers on 64-bit systems, property-based infrastructure, MUD-style chat, and user-controlled theming.
also on Terraform, #software-engineering, #developer-tools
Clopus-Watcher: An autonomous monitoring agent
The author demonstrates Claude Code running in a Kubernetes cronjob as an autonomous on-call agent that monitors pods, applies hotfixes, and documents incidents.
also on #devops, Claude Code, #automation, #developer-tools
How Uber built a software factory for agentic coding: the MCP gateway and the platform underneath
Uber reports that over 70% of pull requests are now agent-written, code output per engineer doubled, driven by a six‑piece platform enabling safe AI‑agent coding at scale.
also on #devops, #automation, #ai-coding, #developer-tools
Coding sessions in Linear – Changelog
Linear Agent can now write and review code using Claude Code and Codex, enabling end-to-end issue-to-ship workflows from within Linear.
also on Sentry, GitHub, Claude Code, #automation, #ai-coding, #developer-tools