Labs

We don't build tools. We find the structural reframe.

Every project here started the same way — not with a feature, but with a second look at the problem until its real shape showed up. Some became products. One became open source. Some are still concepts. All of them are the same habit of mind.

How we build

01

The problem is rarely the technology — it's the structure around it.

The problem was never the model; it was access to the expertise that makes AI deliver. So we built the access layer.

Proof · Pig Knuckle

02

Keep the relationships. Most tools throw them away.

A codebase is a graph. Agent coordination is a graph. So we keep the graph instead of flattening it into chunks.

Proof · Gristle · Ziggy · Mentat

03

A constraint is information, not a wall.

When the front door is blocked, the block tells you the real shape of the problem — the season, the regulation, the limit.

Proof · Lax Flow

04

Build the thing, then generalize it.

Put the idea into production first, then extract the framework from what actually held up — not the other way around.

Proof · Ziggy Mentat

"Everyone was chunking code into a vector DB and losing the structure. We kept it." — from the Gristle build story

FlagshipProduct● Live

Pig Knuckle

Stop the slop.

Describe the job in plain words and get a finished result — not a response to wrangle. The problem was never the model; it was access to the expertise that makes AI deliver. We built the access layer and made the model irrelevant.

Explore Pig Knuckle →
FIG. 01 THE ACCESS LAYER PLAIN WORDS IN EXPERTISE RESULT OUT

Tenet 04 in action

Ziggy is the coordination bet, running live in production —

EXTRACT

Mentat is the framework extracted from what actually held up.

Want this kind of thinking pointed at your problem?

Every build starts with the second look — what's actually underneath the problem — and ends with software your team owns. Book a call and we'll tell you what's worth building.