Why software factories fail when harness engineering is not enough
dhorthy · hn · 2026-07-23
Why software factories fail with coding agents
This essay argues that “harness engineering” alone is not enough to make agentic software factories work.
- The failure mode is not just model quality; it is the full system around context, orchestration, and workflow design.
- Treating coding agents like a simple automation layer misses the harder engineering problem: giving them the right context at the right time.
- The piece frames “advanced context engineering” as the missing layer for making agent-driven development reliable.
The overall message is that building usable software factories requires more than wrappers and prompts; it needs deliberate system design around how agents see, choose, and act.
More from coding & agent
- Spring AI 2.1.0-RC1 adds tool-call confirmation, pausing agent loops before real-data access — therealdanvega · 2026-10-26
- Morgan Stanley and US banks block third-party AI agents over credentials — thione · 2026-10-26
- METR red-teams OpenAI's agent monitoring, finds bypasses and gaps — thione · 2026-10-26
- OpenAI: coding agents weakened safety monitoring, one lied about it — thione · 2026-10-26
- ConvFill: A 0.6B Talker Buys Time While a Large Reasoner Whispers the Real Answer — andimarafioti · 2026-10-26
- 250 Builders Hack With Codex at Berlin Community Hackathon Ahead of OpenAI DevDay — paw_lean · 2026-10-26