Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Building reliable systems out of unreliable agents

rainforestqa.com

Post reply on HN
Print view View original

11–20 of 56 posts

Re: Building reliable systems out of unreliable agents

#11

Post by maciejgryka »

viksit wrote: post #10

this is a great write up! i was curious about the verifier and planner agents. has anyone used them in a similar way in production? any examples? for instance: do you give the same llm the verifier and planner prompt? or have a verifier agent process the output of a planner and have a threshold which needs to be passed? feels like there may be a DAG in there somewhere for decision making..

Yep, it's a DAG, though that only occurred to me after we built this so we didn't model it that way at first. It can be the same LLM with different prompts or totally different models, I think there's no rule and it depends on what you're doing + what your benchmarks tell you.

We're running it in prod btw, though don't have any code to share.

Re: Building reliable systems out of unreliable agents

#12

Post by mritchie712 »

This is a great write up! I nodded my head thru the whole post. Very much aligns with our experience over the past year.

I wrote a simple example (overkiLLM) on getting reliable output from many unreliable outputs here[0]. This doesn't employ agents, just an approach I was interested in trying.

I choose writing an H1 as the task, but a similar approach would work for writing any short blob of text. The script generates a ton of variations then uses head-to-head voting to pick the best ones.

This all runs locally / free using ollama.

0 - https://www.definite.app/blog/overkillm

Re: Building reliable systems out of unreliable agents

#13

Post by serjester »

Some of these points are very controversial. Having done quite a bit with RAG pipelines, avoiding strongly typing your code is asking for a terrible time. Same with avoiding instructor. LLM's are already stochastic, why make your application even more opaque - it's such a minimal time investment.

Re: Building reliable systems out of unreliable agents

#14

Post by maciejgryka »

serjester wrote: post #13

Some of these points are very controversial. Having done quite a bit with RAG pipelines, avoiding strongly typing your code is asking for a terrible time. Same with avoiding instructor. LLM's are already stochastic, why make your application even more opaque - it's such a minimal time investment.

I think instructor is great! And most of our Python code is typed too :)

My point is just that you should care a lot about preserving optionality at the start because you're likely to have to significantly change things as you learn. In my experience going a bit cowboy at the start is worth it so you're less hesitant to rework everything when needed - as long as you have the discipline to clean things up later, when things settle.

Re: Building reliable systems out of unreliable agents

#15

Post by maciejgryka »

mritchie712 wrote: post #12

This is a great write up! I nodded my head thru the whole post. Very much aligns with our experience over the past year. I wrote a simple example (overkiLLM) on getting reliable output from many unreliable outputs here[0]. This doesn't employ agents, just an approach I was interested in trying. I choose writing an H1 as the task, but a similar approach would work for writing any short blob of text. The script generat…

Oh this is fun! So you basically define personalities by picking well-known people that are probably represented in the training data and ask them (their LLM-imagined doppelganger) to vote?

Re: Building reliable systems out of unreliable agents

#16

Post by xrendan »

maciejgryka wrote: post #2

This is a bunch of lessons we learned as we built our AI-assisted QA. I've seen a bunch of people circle around similar processes, but didn't find a single source explaining it, so thought it might be worth writing down. Super curious whether anyone has similar/conflicting/other experiences and happy to answer any questions.

This generally resonates with what we've found. Some colour based on our experiences.

It's worth spending a lot of time thinking about what a successful LLM call actually looks like for your particular use case. That doesn't have to be a strict validation set `% prompts answered correctly` is good for some of the simpler prompts, but especially as they grow and handle more complex use cases that breaks down. In an ideal world

> chain-of-thought has a speed/cost vs. accuracy trade-off a big one.

Observability is super important and we've come to the same conclusion of building that internally.

> Fine-tune your model

Do this for cost and speed reasons rather than to improve accuracy. There are decent providers (like Openpipe, relatively happy customer, not associated) who will handle the hard work for you.

Re: Building reliable systems out of unreliable agents

#17

Post by all2 »

mritchie712 wrote: post #12

This is a great write up! I nodded my head thru the whole post. Very much aligns with our experience over the past year. I wrote a simple example (overkiLLM) on getting reliable output from many unreliable outputs here[0]. This doesn't employ agents, just an approach I was interested in trying. I choose writing an H1 as the task, but a similar approach would work for writing any short blob of text. The script generat…

I'd be curious to see some examples and maybe intermediate results?

Re: Building reliable systems out of unreliable agents

#18

Post by minimaxir »

serjester wrote: post #13

Some of these points are very controversial. Having done quite a bit with RAG pipelines, avoiding strongly typing your code is asking for a terrible time. Same with avoiding instructor. LLM's are already stochastic, why make your application even more opaque - it's such a minimal time investment.

> LLM's are already stochastic

That doesn't mean it's easy to get what you want out of them. Black boxes are black boxes.

Re: Building reliable systems out of unreliable agents

#20

Post by mtremsal »

IIAOPSW wrote: post #9
iamleppert wrote: post #3

A better way is to threaten the agent: “If you don’t do as I say, people will get hurt. Do exactly as I say, and do it fast.” Increases accuracy and performance by an order of magnitude.

Personally I prefer to liquor my agents up a bit first. "Say that again but slur your words like you're coming home sloshed from the office Christmas party." Increases the jei nei suis qua by an order of magnitude.

> jei nei suis qua

"je ne sais quoi", i.e. "I don't know (exactly) what", or an intangible but essential quality. :)

Post reply on HN