We’ve tried. A lot. Custom frameworks and all.

There is really no way to make the ensemble behave with an acceptable level of consistency.

Where we ended up is now having a frontier model generate a whole tree of possible execution plans, and then have the user select one of those path, and then we just run whatever the user chose in a plain sequence until the next decision point that needs user approval.

I've encountered two viable cases: instructions are too complex, too many tools, or wildly different processing steps, in which case it semplify a lot the processing to have a few well defined steps each doing their thing, and a coordinator on top, either sequential, or intelligent, that is only focuesed on next step routing.

the other is memory for conversational retrieval. ai memory is still quite limited, especially if there needs to be a lot of token in context, and context too long impede the ability of llm of focus on the task itself, especially if the context is itself a conversation or a request, so spreading the context along a few agents, and propagating the user request among agent, and having those produce answer fragment for another llm to formulate an answer allows to not lose the conversational context without swamping the llm with noise.

the problem tho remains latency as son as you nest them latency explodes as you can only stream the last layer of llm output

I tried the last crop. Interesting idea but the success rate of any real multi step task always approached 0% the longer it went

I imagine having an agent set up with specific RAG context to solve a specific problem and having another with a different RAG context to solve a different problem can be useful.