What Coding Agents Need From Engineers to Be Trusted
49 min
What do coding agents need from engineers to be trusted? AI consultant Fausto Albers builds production software with agents and doesn't believe anyone still reads the code line by line. Here is what he relies on instead. We cover: Why newer models are "eating the harness", and the shift from context engineering to loop engineeringThe state a coding agent should be in before it runs for hours, and the gates its output has to passObservability with open-source MLflow when you can't read every traceA shared ontology and a three-layer memory stack: a Karpathy-style Markdown wiki in Obsidian, graph memory and a local vector database in DockerDisposable vs durable software, and why software built by intent threatens traditional SaaSWhy AI makes attacks cheaper than defenseThis episode is for software engineers and tech leads who want to give coding agents real autonomy without giving up responsibility for what gets built. Timestamps:00:00:00 - Intro00:00:49 - Why Newer Models Keep Eating the Harness00:02:26 - Will SaaS Survive the Next Ten Years?00:03:39 - From Sociology to Cocktail Bartender to Restaurant Owner00:07:19 - Why Human Preference Can't Be Reduced to Data00:10:10 - What I Learned From a Failed Restaurant Chatbot00:14:16 - Learning Faster With AI and Who Gets Left Behind00:18:38 - What State Should Your Coding Agent Be In?00:22:09 - Building a Shared Understanding Before Kicking Off a Loop00:26:58 - Why Nobody Really Reads AI Code Line by Line00:31:34 - The Real Reason Engineers Push Back on AI00:33:48 - Sloppy, Disposable or Durable: How to Judge Code00:35:18 - A Three-Layer Memory System for Coding Agents00:39:44 - Why Domain Experts Matter When Execution Gets Cheap00:42:23 - How AI Makes Attacks Cheaper Than Defense Guest: Fausto Albers, AI consultant at Wonder Whyhttps://www.linkedin.com/in/stepintoliquid #codingagents #softwareengineering #aiagents
