> It’s actually fine if agents make a lot of boneheaded mistakes. What’s not ok is if they keep making the same mistakes. 

It's simple: when a mistake happens, you run your CAPA process (Google CAPA form and see examples to extrapolate what that process might look like) and determine the root cause and the correction to the process that allowed the mistake to happen in the first place.

(At least as a SaaS vendor in life sciences, when we had a CAPA (e.g. after a SEV0 failure), it would be folded into our SOPs and then we would be required to retrain on the SOP. Auditors would want to see our evidence of CAPAs, the versions of our SOPs, the records of training. All to extreme for most shops, but I add this for context/color)

This is something most eng shops do not have the discipline for since it requires some diligence.

Should it be fully agentic? Should there be human intervention here to approve the CAPA? Open questions to be answered.

The more important question - how would you actually get LLM agents to follow the instructions in your ever-growing CAPA reliably?

It’s all very well having a list of actions to avoid but that doesn’t help if your agents won’t reliably follow it.

I don't see how this could work, right now, since every failing I've had was just generic stupidity of the AI, which is brilliant one minute, and a complete idiot the next.

I think correction actions still require the ability to execute them, which (in all the cases I've had) would require more capable models.

Long term, I think you're probably correct.

That has been a part of post mortems everywhere I work, there was never a name for it though. Just someone would say: “how do we prevent this in the future?” which usually spawns a few tasks that the person who caused the issue or others will pick up.

> Should it be fully agentic? Should there be human intervention here to approve the CAPA? Open questions to be answered.

I think a point we are grappling with is, what necessitates human intervention, like philosophically. Is it accountability? I was thinking about this in terms of code review and it's not like we would fire someone if they broke prod, so at what point do we need someone accountable. Maybe it's for approvals for certain classes of risk (like those associated with actual harm to life). It's also not lost on me that many human systems lack accountability.

Every place I've worked at has positioned their SEV review process as 'blameless' with respect to the engineers who created the code, processes, etc that led to the incident. On paper, it's not about pointing fingers (though I've found this to be somewhat idealistic), but about finding faults in the processes that let bad code slip through the cracks.

It will be interesting to see how the notion of 'responsibility' carries over as agents handle higher and higher levels of abstraction. Agents built using the popular frameworks of today are somewhere in between code, formerly written by engineers, and an actor (like an engineer). When an agent wrote the framework to audit and monitor ops agents who are monitoring the performance and reliability of the code that coding agents wrote, who is responsible for ultimately ensuring it doesn't happen again? Should the findings just be added to whatever RAG and a few prompts/hooks/skills changed by another agent? When does a human follow up and to what degree?

This is the sort of thing that makes me believe that software engineers will never truly go away; at the very least, they serve the very useful function of being ultimately responsible for something going wrong.

This is a general principal in the eng shop I work for, its written into every step, including the SCRUM etc.. What when well, why did it fail, what could be done better. Its all part of root cause analysis.

I personally use it for the agents I use, so why did it fail, what was the root cause, what can we do to prevent it again.

Agreed that domain experts should be making these kinds of decisions, and the right way for an agent mistake to get caught is going to depend a lot on the cost and probability of the mistake. My only point here is that the agent side of the system needs to learn. The software production machine needs to improve over time.

Honestly I feel like I'm taking crazy pills these days.

Write a program! If it has a bug, debug it and update the program. Use an LLM to debug it if you must. If you're doing something repeatable, then use a traditional automation scheme.

If an LLM has to handle a subsection of the program, so be it. Just wrap that part with clean inputs and outputs.

Why does it feel like we've forgotten how to automate things?