A private residency.

Six weeks to learn how to use AI on the work where your judgment is the product.

A private residency is six weekly one-on-one sessions with Bud Bhattacharyya, founder of re:compound, for senior leaders carrying complex, consequential work.

Not a course.

Not prompt training.

Not an abstract AI use case.

You bring the live work: the deal that keeps moving, the portfolio company where everything is shifting at once, the strategic decision waiting on your call, the organizational problem nobody is framing cleanly, the board question you keep returning to, the context you lose every time your attention is pulled somewhere else.

We use that work to build a new way of collaborating with AI — one that compounds your judgment instead of simply producing more output around it.

The problem

AI is already good at producing more.

More summaries. More drafts. More research. More analysis. More options.

But for senior leaders, more output is rarely the actual bottleneck.

The expensive thing is the state of understanding you are carrying in your head:

• the half-formed read on the situation;
• the assumption that quietly shifted;
• the client dynamic that changed;
• the board member’s concern that was not quite said;
• the team tension that makes the obvious answer wrong;
• the unresolved call you keep returning to;
• the strategic question that is too complex to hold all at once.

That state is valuable. It is also fragile.

Your calendar breaks it apart. Your meetings interrupt it. Your team sends more material into it.

Your attention moves elsewhere, and when you come back, you are not simply continuing the work. You are rebuilding the state that made the work intelligible.

From the outside, it looks like you are working on the problem. Often, you are catching back up to where you already were.

Using AI to generate more output below you does not solve that. In many cases, it makes the problem worse. More artifacts roll up to the same scarce judgment.

The real opportunity is different.

What changes

You learn to use AI not as a faster producer of artifacts, but as a governed cognitive partner for complex work.

The aim is to help you hold, sharpen, and return to the live state of a consequential problem:

• what is actually going on;
• what has changed;
• what remains unresolved;
• what matters now;
• what branch should be kept alive;
• what needs to be dropped;
• what the next real move should be.

This is a different way of working.

Not: ask the model, get the answer.

Not: delegate the draft and edit the output.

Not: generate more options and decide alone.

Instead, you learn how to govern a live human-AI work loop: how to frame the problem, build the field, pressure-test the model, preserve exactness, avoid polished wrongness, and turn the work into checkpoints you can actually re-enter.

By the end of six weeks, AI should work differently for you.

It should help you return to complex situations without starting cold.

It should help you reach for problems you were avoiding because they were too hard to carry all at once.

It should help you multiply what your judgment can do.

Not replace it. Not route around it. Not bury it under more output. Compound it.

What we work on

The right way to learn this is through real work.

Each week, you bring whatever is most live, consequential, and complex. It might be:

• a deal that will not sit still;
• a portfolio company question where every option has a cost;
• a founder or CEO decision with no clean answer;
• a strategy problem that keeps generating more analysis but not more clarity;
• a senior team issue where the stated problem is not the real problem;
• a board, investor, or client conversation you need to prepare for;
• an AI transformation question your organization keeps treating as a tool problem when it is really a judgment problem.

It does not have to be the same situation every week. In many cases, it should not be.

The object is whatever is most in front of you that week.

The discipline is choosing the object before the session and bringing something real enough that your judgment matters.

What happens in the sessions

Each session is a 90-minute private working session.

We do not begin with a curriculum. We begin with the live matter.

Together, we use AI to:

• define the real problem under the stated problem;
• build and update the live state of the situation;
• separate signal from noise;
• preserve the distinctions that matter;
• test branches without prematurely closing them;
• identify where your judgment is actually needed;
• create a usable checkpoint so the work can be resumed later without rebuilding from scratch.

Along the way, you learn the method by using it.

You learn how to collaborate with AI on consequential work without collapsing into generic prompting, output review, or polished answer-shapedness.

The session should move the actual problem. But the larger purpose is that you learn how to run this kind of loop yourself.

What you take with you

The durable product is not six calls.

It is a way of working.

You leave with a practical method for using AI on the work where senior judgment is the actual thing being exercised:

• how to build the field before asking for the artifact;
• how to keep important branches alive without drowning in them;
• how to use AI to sharpen judgment rather than substitute for it;
• how to tell when the model is producing a nearby answer instead of the real one;
• how to preserve the state of a complex matter so you can re-enter it later;
• how to turn live thinking into a checkpoint, memo, decision frame, or next move.

This is not dependency on me being in the room.

It is a residency: a concentrated period of coached practice on your real work, designed to leave you with a method you can keep using.

The one condition

This is only worth doing if there is real work worthy of the attention.

Not a hypothetical AI use case. Not a general wish to “get better at AI.” Not curiosity detached from a consequential situation.

The work has to be live.

Your judgment has to matter.

The cost of staying stuck, missing the real issue, or rebuilding the same context every week has to be high enough to justify serious attention.

That is the condition.

Format

Six weekly sessions. Ninety minutes each. One-on-one with Bud. Conducted by Zoom.

Built around whatever consequential work is most live for you that week.

$6,000.

Flat fee.

A second seat is available for $1,500 additional.

The second seat is for the person who helps you carry the situation alongside you: a chief of staff, deputy, operating lead, or trusted colleague.

Optional, but often useful.

If the residency leads to a longer advisory engagement, the fee is credited toward that engagement.

If this sounds like it might fit

Reach out for a 30-minute fit call.

Bring one real situation.

The call is for both of us to see whether the method can move something that matters.

No deck. No pitch theater. No chasing.

If it fits, we book the first session. If it does not, we will both know quickly.

—Bud

Book time with Bud

Or reach out to elissa@recompound.ai.