---
title: "AI Harness Engineering Guides"
description: "Build reliable coding-agent workflows with clear repository, evaluation, and delivery practices."
canonical_url: https://optimi.com/en/guides/tags/ai-harness-engineering
md_url: https://optimi.com/en/guides/tags/ai-harness-engineering.md
---

# AI Harness Engineering Guides

Build reliable coding-agent workflows with clear repository, evaluation, and delivery practices.

- [AI Harness Engineering](/en/guides/harness-engineering-coding-agents.md): AI harness engineering designs a bounded, testable coding-agent loop with task contracts, repository context, independent controls, and reviewable delivery evidence.

- [Agent-Navigable Repositories](/en/guides/agent-navigable-repositories.md): Build agent-navigable repositories with layered instructions, a compact repository map, runnable checks, and durable handoffs that coding agents and reviewers can both trust.

- [Secure Agent Sandboxes](/en/guides/secure-coding-agent-sandboxes.md): Harden coding agent sandboxes with ephemeral credentials for agents, scoped identity, controlled egress, agent execution isolation, audit evidence, and safe recovery from policy denials.

- [Scaling Agent-First Delivery](/en/guides/scaling-agent-first-delivery.md): Scale agent-first software delivery with task claims, isolated CI capacity, review-aware queues, deterministic gates, and measured autonomy.

- [Agent Evaluation and Observability](/en/guides/agent-evaluation-observability.md): A practical framework for coding agent evaluation: versioned cases, deterministic graders, multi-trial variance, privacy-aware agent observability, and redaction proof.