Engineering · 18 Aug 2026

Memory is not a search box

Giving an AI everything your company knows is the easy version of the problem, and the wrong one. Scope is what makes memory usable.

The demo everyone builds first is a search box over the company drive. Ask a question, get an answer, watch the room nod. It is a good demo. It is not a coworker, and the gap between the two is the interesting part.

#Two things that get called memory

Retrieval

Finding the document that answers a question. It is a solved shape: index the files, embed the chunks, rank the results.

Memory

Knowing that Northwind always pays thirty days late, that Marco signs anything over five thousand, and that the discount you gave Halcyon in March was a one-off nobody should repeat.

Retrieval finds a page. Memory carries a rule. Nobody wrote the Northwind rule down in a document called Northwind Rules. Somebody said it in a thread, once, eleven months ago, and everyone who matters has known it since.

#Scope is the feature

The instinct is to give the AI everything, because more context sounds like more intelligence. In a company it is the opposite. A finance coworker that can read the recruitment folder does not become better at chasing invoices. It becomes a question your general counsel has to answer.

So a coworker in Alfera reads the sources attached to it and nothing else. Not "everything, filtered at query time" — the difference is the whole design:

What most products do

Everything, filtered on the way out

The coworker can reach the whole drive. A rule decides what to show for this question, for this person, this time.

The filter is a permission you have to trust the query to apply. Every new kind of question is a new chance to apply it wrongly.

What we do

Only what is attached

The coworker can reach the ledger and the contracts. The recruitment folder is not on the list.

There is no query that reaches it, because there is nothing to filter. It is not hidden from the coworker. It is not there.

Both answer the same question. Only one of them can be wrong in a way you find out about later.

That has a consequence worth stating plainly, because it is the part people check: change the scope and it forgets. Remove the ledger from a coworker and the facts it drew from the ledger stop being available in the same second. It does not keep a summary of what it used to know.

A summary of a document you are no longer allowed to read is a copy of a document you are no longer allowed to read.

#Where the rules come from

The Northwind rule is not in a file, so it cannot be retrieved. It gets learned the way a new starter learns it — somebody corrects them.

Don't chase Northwind, they always pay late and they always pay.

That sentence, in a thread, is worth more than the whole payments folder. So we treat a correction as the highest-value thing that happens in a workspace.

Somebody corrects itOne sentence, in the thread, where they already are
It becomes a ruleAttached to the thing it corrects, carrying who and when
It shows up cited“I skipped Northwind, as you asked” — with the link
A correction is not a message that gets read once. It becomes a rule with a link back to the moment somebody said it.

Two reasons the citation matters more than it seems:

  1. You can check it. A rule you cannot trace is a rule you have to take on faith, and nobody gives an AI that.
  2. You can kill it. When the rule stops being true, you correct it once, in the same place, and it stops being applied.

#What we got wrong

We built the eager version first: a compiler that read everything overnight and produced a tidy summary of the company. It was elegant and it made things worse.

Two findings from killing it, both uncomfortable:

  • A cached summary that nobody invalidates is poison. It answers confidently from a world that has moved on. Measured against reading the sources at the time of the question, it was worse, not slower-but-equal.
  • Most of what a company writes is not knowledge. The overwhelming majority of the raw material was activity — who did what, when. Compiling it produced volume, not understanding.

What survived was smaller and duller: read at the time of asking, inside the scope, and let corrections accumulate as rules with links back to where they came from.

#The line we hold

A coworker should be able to explain any answer with two things: the source it read, and the correction it applied. If it cannot produce both, we would rather it said it does not know.

That sounds like a limitation. In practice it is the reason people give it the second job.

Questions people ask after this

What is the difference between retrieval and memory?

Retrieval finds the document that answers a question. Memory carries a rule that nobody wrote in a document — that this customer always pays late, that this person signs anything over a limit. Retrieval finds a page; memory applies a rule.

Does a coworker read everything my company has?

No. It reads the sources attached to it and nothing else. Anything not attached is not in its world at all, rather than being filtered out at the moment somebody asks.

What happens when I remove a source from a coworker?

It forgets, in the same second. The facts it drew from that source stop being available, and it keeps no summary of what it used to know.

How does a coworker learn a rule that is not written down anywhere?

Somebody corrects it, in a sentence, where they already are. The correction becomes a rule attached to the thing it corrects, carrying who said it and when, and it shows up as a citation the next time the coworker acts on it.

Oliviero Pinotti

Founder

ShareXLinkedIn

Put one on a real job this afternoon

Connect one tool and give it one task. You will know inside ten minutes.