Ask HN: Are you building an AI company brain?

Hi, the idea is good but the implementation is quite hard. And yes, we have build such system and it is not very easy, it is not only about saving data but allow any agent to find it, easily, without costing too much. With time, you will have more and more information in your brain, the difficulties you will find: 1 - structure the information correctly 2 - create relationships between related content 3 - maintain such information when something change 4 - searching for a keyword but a same concept can be expressed in several way, words, ... 5 - find the information in all the accumulated content, can be GB of data to search for... 6 - and how the system react when it does not find the information, is hallucinates... both for a search and for adding/updating information 7 - more and more complex problems...

So, trust me, it is way more complex to do, an AI model is not do to analyse big data, even LLM have limits... the hardest part is to keep this information structured, correct and well organized at the speed of AI, without costing too much. The risk is to create more and more false information and the system become unmaintable... or you spend your life on it.

One trick, if you want to do something like that, create a node based graph knowledge base, it is just the start. Noticed that we are looking to release our brain db (working on it since a year now... it is amazing, but complex to exploit it in production), but I do not know when... the plan is not clear, we use it internaly only for now.

Thanks, I'll keep an eye on that. I think that's important to store the raw data somewhere so you have the flexibility to "rewind" if things get out of control or start to degrade. Maybe adding some evals could also help.

I have built similar solutions for multiple clients.

Some basic tips I hope that help you.

1. When the system is up and running don't run agents in a loop, wake them up when something happens. An email arrives, something changed, a timer triggered. There is no reason for 24x7 agents most of the time.

2. You want a system that does not allow conflicts, as your baseline, for projects where many people are working on the same thing. Think git for documents. How is not important, it's important that it's there

3. A basic coding harness like Claude Code is going to do a million times better than anything you can build yourself. It's already optimized for dealing with tasks in a complex environment and gets updated near daily. Don't roll something yourself or buy a special solution from a vendor

4. It's super easy to provide context for a meeting. It's very hard to get people to read them. Spend time making sure pre-reads are short, concise and relevant.

5. Only have one source of truth for core data. An AI agent will happily find an outdated document and assume that's the truth. Make sure it's strongly enforced that important structured data is in an easy to search single database.

Build it incrementally. Pester your CTO about what task is the most important then start there. Spend your time collecting feedback from the people who use the system and build as little as possible. Very easy to add every feature, harder to make sure you only build the features that make the system better.

When that is said, this could easily become a very fun project to work on! Feel free to reach out if you get blocked.

Thanks! It helps a lot! If I merge 1, 2 and 3 that would mean creating a service API with the Claude SDK behind to handle requests whenever some employee wants to make a read/write operation.

The write operations will be made only for the documents included in the repo if the agent allows it according to user permissions. Am I right or did I get something wrong?

I'm also thinking about "real time data" like google docs, chats, transcripts, databases, logs...

My current idea would be to keep that out of this repo and have each user to query them (if they have permissions) with their own personal account. This way the system is more simple and maybe cheaper because each employee can use their current Agent plan instead of the service API tokens that are way more expensive and difficult to limit.

What do you think about this?

Oh man, that is a rough space to be in. Don't get me wrong, I am kinda jelly of your assignment here, but there are bunch of things to keep in mind.

"some sort of entity that keeps the context of what's going on in the company"

This is the killer piece, because I know what it would mean to me, but did CTO give you a clear indicator of what he means? All the meetings, drafts and whathaveyou won't automatically capture those.

- wiki equivalent is good, but it comes with the expectation that it is updated, current and accurate ( with as little as possible not listed -- which can be a pickle for some uses ) - databases - you want to be careful here for obvious reasons esp if your databases hold anything beyond simple purchase orders - 24x7 agent could cost less long term if you go for local inference ( and accept some speed trade-offs ) - orchestration - there is a lot money sloshing around this one including big boys - I did not see good corporate orchestration suite yet ( though I am sure it exists )

<< is there a better way to accomplish this? Am I over complicating things? Am I over simplifying things?

Maybe. Get the feel whether CTO is actually asking for a prettier dashboard or something more meaningful. I am not joking.

The CTO wants it, there's no way around that :).

I expect frontier labs to solve this. Anthropic has a recipe here: https://platform.claude.com/cookbook/capabilities-knowledge-...

When I recently spoke to their engineers about this, they mentioned graph maintenance is very expensive and would likely be cost prohibitive for most orgs.

Thanks, will check out the article.

For the first step, you should store docs, pages and other text based data using advanced RAG in a vector db, then you can use LangGraph to do any kind of workflow with it. Know some good open-source packages helping you manage them

To what kind of workflows do you thinks LangGraph would help in this domain?

Did you look into open source projects covering for these needs? There are many out there.

If so, why didn't you choose any of them?

I can tell you our experience, but I'm prob biased.

Also, boca o river?

I've been looking at several projects but didn't find something that convinced me, most of what I've seen are more focused on software generation "AI factories".

- This is interesting but more complex https://github.com/langchain-ai/design-agent. - This talk points to something similar but ingesting all the data into the system and chunking it to allow semantic search later https://ai.engineer/talks/KWmkMV0FNwQ-building-alices-brain-...

Plase tell your experience (I'll take a look at https://github.com/desplega-ai/agent-swarm/)

"El más grande sigue siendo River plate"

Barcelona or Real Madrid?

Boca XD. Igual no sigo mucho ya.

Tbh, with the swarm, we cover for that particular need. The caveat is that we don't try to be the SOT. At the end of the day, your company knowledge is distributed by design, so you're system needs to account for that. Instead of trying to have everything up to date, you need to remember where to look for the knowledge... even if it means burning a bit more tokens.

Yeah, we dont have a SOT but we're trying to at least hace something close to that. Nowadays theres siloed data and no one has access to the whole picture.

Happy to talk through the options, swar.am/memory is how it looks like for the swarm.