Work Method Foundry Field Notes About Edge Brief
← All work

Selected work · founder systems · venture

Rehearsing the room that matters.

An AI rehearsal adversary for founders — the hard questions asked early, and a story that leaves every session as a file, not a feeling.

Anvil SessionStoryHistory REHEARSAL · LIVE
The session
you“We help teams ship faster with AI.”
anvilWho feels that pain first — and what do they pay for it today?
you“Engineering leaders, mostly…”
anvil“Mostly” is where rooms say no. Pick one, and tell me their Tuesday.
story.json · compiling
buyer"VP Engineering"
frame"cycle time"
first_dollar"paid pilot"
weakest_claim"10x productivity"

The engagement

An accelerator asked us to look into a pattern that was costing its portfolio real money: its most promising founders kept losing the rooms that mattered. The pitches had been rehearsed — on partners, on mentors, on friends — and everyone had nodded. Then the term-sheet meeting or the enterprise sales call arrived, asked the one question nobody had rehearsed, and the room was gone.

The engagement began with an Edge Brief, then a data room of cohort histories: which founders advanced, which stalled, and what the post-mortems blamed. Our agents read the record against what actually happens in funding and sales conversations, and the ledger put a name on the gap — practice audiences are polite, and politeness is a defect in a rehearsal partner.

The process

Discover → Architect → Build → Commercialize.

Discover
The pattern held across cohorts: outcomes tracked less with the quality of the idea than with whether the founder had already faced the hardest version of the room’s questions. Almost none had. The scarce resource was not advice; it was honest pressure, available on demand.
Architect
Candidates included more coaching hours, question banks, and pitch templates. The matrix killed the comfortable options for the same reason the problem existed — they added more politeness. The survivor paired an AI adversary that asks what the room will ask with a compiler that turns every session into a structured artifact.
Build
The system was built and pressure-tested against the record: sessions run on pitches whose real-room outcomes were already known. The questions that predicted failure earned a permanent place in the adversary’s repertoire.
Commercialize
The tool went into use with founders — and the screening showed the rehearsal gap is universal wherever high-stakes rooms exist, which is why it now operates as Anvil, a Modven venture. One problem, paid twice.

What the process selected · 01

THE ADVERSARY
anvilYour best customer churns tomorrow. What did you miss?
anvilWhy hasn’t the incumbent already built this?
anvilWhich number in your deck would a diligence team attack first?
questions harvested from real rooms · asked before the room can

An audience that refuses to be polite.

This candidate won selection because it supplies the one thing a founder’s network cannot: pressure without consequence. The adversary’s questions are harvested from what real rooms actually ask — diligence attacks, churn scenarios, the incumbent question — and it asks them before the room can, when the answer can still be fixed. Selection favored discomfort on purpose, because discomfort was precisely what the polite rehearsals lacked.

What the process selected · 02

Every session compiles to a file, not a feeling.

The second survivor answers what happens after the rehearsal. Confidence evaporates; artifacts do not. Each session compiles into a structured brief — the buyer, the frame, the first dollar, the claims that survived and the ones that did not — so preparation accumulates across sessions instead of resetting with each burst of adrenaline. The file is the founder’s story under version control.

story.json · the artifact
buyer"VP Engineering"
pain_today"6-week release cycle"
frame"cycle time, not headcount"
first_dollar"$15K pilot, 60 days"
objections_survived7 of 9
every session ends in a file, not a feeling

Where it stands

Live with founders, and now a Modven venture. The engagement diagnosed one accelerator’s pattern; the screen showed the rehearsal gap belongs to everyone who walks into high-stakes rooms, and Anvil is the second harvest. Its rule stays fixed: the rehearsal should be harder than the room.

Next case: Kiss →

Your business is holding problems worth money. Let’s find them.

Every engagement begins the same way: one paid conversation, one written brief, one clear answer about the edge worth building.

See more work