Claude Code @anthropicai
A software factory is a repeatable loop around software work. If you're building a software factory, code good enough to ship still needs human taste and ownership. We'll discuss this including
For much of human history, we've evaluated code quality via code review: someone reads what you wrote and makes sure it's clean, thoughtful, fast, understandable, and tests well. For agents, that
Taste is recognizing quality and greatness in the absence of all the evidence. Judgment is committing to it in the presence of some risk. It’s the ultimate foundation of great work. When you look at a
A software factory is harnessed loops at scale. You can run the loop with humans in it (light factory): trading judgment and concentration against speed and breakage. Or you can ignore the humans
Taste used to be a byproduct of the reps. Agents took the reps. So if you're junior you now have to go get the taste (and judgment) on purpose. As I was building my career, I learned to think about
In the past year, the conversation around agentic engineering has moved to harnesses and loops, fleets and software factories. My 2c is engineers need to own the outer loop - the accountability for
In most conversations about agentic engineering, the action has changed from prompting to operating. Here's a frontier looking into the fog: software factories, goals, loops, background sessions,
Web developers often treat the browser as a black box that magically transforms HTML, CSS, and JavaScript into interactive web applications. In truth, a modern web browser like Chrome (Chromium),
Coding agents are extraordinarily good now and getting better fast. The interesting consequence is that the hard part of engineering moved from writing code to deciding whether to trust it, which
Loop engineering is replacing yourself as the person who prompts the agent. You design the system that does it instead. A loop here can be thought of a recursive goal where you define a purpose and
Starting more AI agents is easy now. However, more agents running doesn’t mean there's more of you available - your cognitive bandwidth doesn’t parallelize. All the judgement to actually steer them
Right now, it's too easy to let AI write the code while you skip the learning. The bug gets fixed but your mental model doesn't move. It might get worse over time. We are silently trading future
