Models know facts. Experts know what to do with them.

Professionals publish how they decide, so the AI you already use can read one expert’s real past decisions before it answers you.

Browse Decision Models Publish yours

Free during beta · no account · runs on the AI subscription you already pay for.

Works with

Claude app
ChatGPT
Cursor
Claude Code
Gemini CLI
and more

Stop teaching every AI the same judgment from scratch

A rules file.
A paragraph of context, pasted into every new session.
One more correction, when it confidently suggests something you would never do.

It works until you hit a call you have not made before. Then you are writing down a guess — and your AI follows it to the letter.

Rules tell an agent what to do. They rarely tell it what to put first when the facts are incomplete and two reasonable goals pull against each other. A Decision Model gives your AI an approved record of how one professional makes exactly those trade-offs — what they weigh, where they stop, which way they go when the two collide — and your AI applies it to your situation, including the ones they never saw.

One question, already answered

You ask your own AI, in your own tool. It consults the model first and answers with that judgment already in hand.

Example inside your AI tool

You

We're a 20-person Series A. Our marketing site looks dated next to our two biggest competitors and the board keeps bringing it up. Do we redesign the brand or the site first?

Consulting Ines’s Decision Model

Your AI

Recommendation: The site, and only the site. Keep the name, the mark and the wordmark exactly where they are. New typography, new photography, and a homepage headline that says what you do and for whom.
Why: “Dated” is almost never an identity problem. What a board is reacting to is the typography and the fact that your headline does not separate you from the two companies beside it — both of which live in the pages, not in the mark. Identity is the output of positioning: change it before the positioning is settled and you will change it again, having spent the one design budget you had on the wrong problem.
What would change this: If the company has actually moved — new segment, new buyer, a different room — then the mark is fighting the new positioning and the brand goes first, site second. That is a change in audience, not a change in taste, and it is the only version of this question where the logo is the answer.
Relevant precedent: Seed-stage B2B startup, 11 people, 40 paying customers, no clear positioning yet. The founder wanted a full rebrand — new name, new logo, new site — before the next raise. → No rebrand. Fix the homepage headline and the product's empty states; leave the name and mark alone.
Outside this model: Nothing here about what a redesign should cost, who should build it, or how to run the agency relationship. This model covers what to change and what to leave alone.

Ines’s own answer, on a question they chose. Your AI writes the wording for your situation; the judgment is theirs.

Ines is a reference model — a composite avva wrote, not a real person. A real expert’s page carries their own answer.

Two AIs, and neither of them is ours

You keep the AI you already use. So does the expert. Between them sits avva, holding what they decided.

Which is why there is no second subscription and no per-call charge, and why a better model on your side makes the advice better.

The expert’s AI
reads the work history they already have, and drafts a model. They approve every decision in it before anything is submitted.
avva
stores and serves the finished model. It writes no answers, and it keeps none of the source material the model was built from.
Your AI
reads it before it answers you, on your own subscription, in the tools you already work in — and names where the expert had nothing to say.

When you would reach for an expert

The calls where a wrong answer is expensive, and the reason matters more than the answer.

Before you commit to an approach

Nobody senior is around to check it. Someone who has made this call tells you what they would not sign off on.

When the room disagrees

Two good options and no tie-breaker. You see how one professional actually settled it, and why.

When you are new to the problem

Not the average of the internet. You get the questions they ask first, before anyone has an opinion.

For companies

Ask your seniors without booking them

The same few people get pulled into every hard call. Put how they decide on a shelf of your own, and every agent your team runs can ask them first.

Start a private workspace

The one who prices the work, the one who sets the standard, the one whose no is final — each publishes the calls they actually made, onto a shelf only invited colleagues can open. Their standards reach every review, and they stop repeating themselves in each one.

Internal means internal. Never in the public catalog, never in search — to anyone outside, your shelf answers as if it does not exist. Remove someone and their access stops on their next call.

In the marketplace

One professional per model — their real past decisions, with the reasons on the record.

See the marketplace

For experts

Be the expert their AI asks

You answer the same question over and over, and each answer helps one person once. Publish once and it becomes a link you hand a client — their AI works to your standards, without you in the room.

Soon Earn from it when paid access opens — and it stays your call whether your model is ever paid.

Build your Decision Model What it takes

Your own assistant does the reading, on your machine and your account. It shows you titles first and you strike anything private before a line of text is submitted. avva stores the finished model, never the decisions it came from.

What this is not

Five things people assume on first read, and the honest answer to each.

Not a chatbot of the person. No clone and no impersonation. It is a published record, read by your own AI.
Not the person answering. Nobody is notified, nobody is on call, and nothing you ask reaches a human.
Not an AI we run. A connection returns their published Decision Model and a short procedure. Your own AI does the thinking.
Not an accuracy claim. Coverage labels describe breadth, nothing more. We publish no score that predicts how often someone is right.
Not a file you download. Your AI reads it live, request by request, and we count the requests. Nothing to export, nothing to resell.

Nothing an expert did not approve gets published. Client names, employer names, and source quotes are stripped from what does. Details in the Privacy Center. Free during beta.