<![CDATA[The Next Commit]]>The real stories behind AI driven development, straight from the people building it.

Every week, we sit down with developers, engineering leaders, and AI transformation champions who are actually using these tools on the ground. It's a grounded conversation about what's working, what's failing, and what's next.

In a moment when every tool claims to be a game changer, this is where builders talk to builders, sharing the patterns, pitfalls, and hard won wisdom that only come from doing the work.

]]>https://riverside.comRiverside.fm (https://riverside.com)Wed, 09 Sep 2026 20:07:43 GMTThu, 16 Jul 2026 16:00:44 GMT60Sonar<p>The real stories behind AI driven development, straight from the people building it.</p><p></p><p>Every week, we sit down with developers, engineering leaders, and AI transformation champions who are actually using these tools on the ground. It's a grounded conversation about what's working, what's failing, and what's next.</p><p></p><p>In a moment when every tool claims to be a game changer, this is where builders talk to builders, sharing the patterns, pitfalls, and hard won wisdom that only come from doing the work.</p>episodicSonarthenextcommit@sonarsource.comno<![CDATA[Compose the work. Conduct the flow. Let the agents play.]]>Ian Johnson’s team ships six to eight small tickets per developer each day. The real story is the system behind that speed. Ian explains how precise requirements, a truthful project charter, small pull requests and layered verification make coding agents dependable. We discuss comprehension debt, legacy refactoring, the amplification thesis and why the developer’s emerging role is to build the thing that builds the thing.

About Ian

Ian Johnson is a staff engineer at Parento and author of Harness Engineering, a practical guide to reliable workflows for non-deterministic agents. His work explores how charters, verification and feedback loops can increase delivery speed without surrendering quality or accountability.

In this episode

  • 01:14: Introducing Ian and Harness Engineering
  • 02:16: Ian’s role at Parento and the book’s origins
  • 03:10: From autocomplete to agentic coding
  • 04:42: Claude Code, Codex, Pi and parallel agents
  • 06:34: Engineering a system for parallel agents
  • 08:08: Review agents, checks and human accountability
  • 10:01: Keeping changes small enough to review
  • 11:01: One Jira card, one pull request
  • 12:23: Falsifiable criteria and explicit exclusions
  • 13:28: AI-assisted refinement without invented requirements
  • 14:48: Decomposing features into releasable work
  • 17:01: Shipping six to eight cards per developer daily
  • 17:59: Feeding review lessons back into the harness
  • 20:20: The charter as an agreement with agents
  • 23:17: Agents amplify order and disorder
  • 24:55: Structuring and indexing project context
  • 26:32: Why inaccurate rules damage output
  • 27:58: Truthful rules and living migration plans
  • 29:39: Refactoring legacy code with tests and TDD
  • 31:58: 100% coverage as a local safety boundary
  • 33:10: Why cheap code must still be maintainable
  • 35:47: Human oversight and comprehension debt
  • 36:57: Resisting cognitive surrender
  • 39:33: Pairing a junior, senior and coding agent
  • 42:57: Evaluating charter and harness changes
  • 45:56: Raising the developer’s abstraction level
  • 47:33: Building the thing that builds the thing
  • 51:06: Amplification and acceptable reliability
  • 53:19: Why a hook beats a rule

Key ideas

  • Tickets control what an agent builds; the charter controls how it builds it.
  • Small batches and falsifiable acceptance criteria keep agent work reviewable.
  • Tests, hooks and analysis turn standards into evidence.
  • Agents amplify their environment, but humans retain accountability.

People and links

]]>409f29dd-60b0-411e-9caf-4391413e4a59Tue, 08 Sep 2026 10:00:00 GMT<p>Ian Johnson’s team ships six to eight small tickets per developer each day. The real story is the system behind that speed. Ian explains how precise requirements, a truthful project charter, small pull requests and layered verification make coding agents dependable. We discuss comprehension debt, legacy refactoring, the amplification thesis and why the developer’s emerging role is to build the thing that builds the thing.</p><h2>About Ian</h2><p>Ian Johnson is a staff engineer at Parento and author of <i>Harness Engineering</i>, a practical guide to reliable workflows for non-deterministic agents. His work explores how charters, verification and feedback loops can increase delivery speed without surrendering quality or accountability.</p><h2>In this episode</h2><ul><li>01:14: Introducing Ian and <i>Harness Engineering</i></li><li>02:16: Ian’s role at Parento and the book’s origins</li><li>03:10: From autocomplete to agentic coding</li><li>04:42: Claude Code, Codex, Pi and parallel agents</li><li>06:34: Engineering a system for parallel agents</li><li>08:08: Review agents, checks and human accountability</li><li>10:01: Keeping changes small enough to review</li><li>11:01: One Jira card, one pull request</li><li>12:23: Falsifiable criteria and explicit exclusions</li><li>13:28: AI-assisted refinement without invented requirements</li><li>14:48: Decomposing features into releasable work</li><li>17:01: Shipping six to eight cards per developer daily</li><li>17:59: Feeding review lessons back into the harness</li><li>20:20: The charter as an agreement with agents</li><li>23:17: Agents amplify order and disorder</li><li>24:55: Structuring and indexing project context</li><li>26:32: Why inaccurate rules damage output</li><li>27:58: Truthful rules and living migration plans</li><li>29:39: Refactoring legacy code with tests and TDD</li><li>31:58: 100% coverage as a local safety boundary</li><li>33:10: Why cheap code must still be maintainable</li><li>35:47: Human oversight and comprehension debt</li><li>36:57: Resisting cognitive surrender</li><li>39:33: Pairing a junior, senior and coding agent</li><li>42:57: Evaluating charter and harness changes</li><li>45:56: Raising the developer’s abstraction level</li><li>47:33: Building the thing that builds the thing</li><li>51:06: Amplification and acceptable reliability</li><li>53:19: Why a hook beats a rule</li></ul><h2>Key ideas</h2><ul><li>Tickets control what an agent builds; the charter controls how it builds it.</li><li>Small batches and falsifiable acceptance criteria keep agent work reviewable.</li><li>Tests, hooks and analysis turn standards into evidence.</li><li>Agents amplify their environment, but humans retain accountability.</li></ul><h2>People and links</h2><ul><li><a rel="noopener noreferrer nofollow" href="https://www.linkedin.com/in/tacoda" target="_blank">Ian Johnson on LinkedIn</a></li><li><a rel="noopener noreferrer nofollow" href="https://medium.com/@tacoda" target="_blank">Ian Johnson on Medium</a></li><li><a rel="noopener noreferrer nofollow" href="https://dev.to/tacoda" target="_blank">Ian Johnson on DEV Community</a></li><li><a rel="noopener noreferrer nofollow" href="https://leanpub.com/harness-engineering" target="_blank"><i>Harness Engineering</i> on Leanpub</a></li><li><a rel="noopener noreferrer nofollow" href="https://www.parentoleave.com/" target="_blank">Parento</a></li></ul>no00:51:5311Compose the work. Conduct the flow. Let the agents play.full