MANIFESTO

Why we built Originalis.

01 · THE WORK

The work

Venture capital refuses to be systemized. It is unstructured, nonlinear, and deeply human. It runs on ambiguous signals, tribal knowledge, incomplete information, and narrative as much as data. Judgment is crafted, not inherited. Conviction forms through context, not formulas. Every other corner of private capital operates on standardized workflows and repeatable processes. Venture does not, and that is not a flaw in the asset class. It is the nature of the work, a feature not a bug.

I have spent two decades inside it, as a founder, as a general partner, as an LP, as an operator. I have lived the contradictions and the nonlinear judgment that define this craft. The software built for it never has. Every tool that touched venture either forced structure onto work that does not hold it, or fled to the parts of private markets where structure already exists. Venture was left to run on memory, hacked systems, and heroics.

THE REST OF PRIVATE CAPITALVENTUREBUYOUTPRIVATE CREDITREAL ASSETSGROWTH EQUITYAMBIGUOUS SIGNALREVERSALDOUBTCONVICTIONSTANDARDIZED WORKFLOWS ON ONE SIDE. THE WORK ITSELF ON THE OTHER. A FEATURE, NOT A BUG.
FIG. 01 · THE SHAPE OF THE WORK
02 · THE JUDGMENT TAX

The judgment tax

The cost of that is not on the P&L. It is the hours lost switching between nine tools that do not speak to each other, none of them crafted natively for venture. It is the same founder meeting done twice by two associates who did not know the other had started. It is the institutional memory that walks out the door when someone leaves. It is the feedback loop where judgment forms in a Monday partner meeting and is forgotten by Thursday. It is the cognitive load of holding the whole firm in your head because no system can hold it for you.

I call it the judgment tax. The judgment itself is not the problem. The cost is everything around it: writing it down so it can be used again, socializing it so the room shares it, convincing partners who were not in the meeting, and keeping it in the firm after the people who formed it move on. It taxes speed, it taxes conviction, it taxes coordination, and it quietly shapes every outcome. The deepest cut is what it does to learning. Investors develop judgment through repetition. Patterns emerge, taste forms, preferences sharpen. But the software forgets the moment a deal ends, so every new opportunity starts from zero. Analysts redo the same work. Partners restate the same instincts. Intelligence resets when it should compound.

WHAT THE FIRM LEARNSDEAL ONEDEAL TWODEAL THREEDEAL FOURDEAL FIVEJUDGMENT, KEPTJUDGMENT, RESET AT EVERY DEAL ENDTHE TAXINTELLIGENCE RESETS WHEN IT SHOULD COMPOUND. THE GAP IS THE TAX.
FIG. 02 · THE JUDGMENT TAX
03 · THE BELIEF

The belief

Judgment compounds only if it is kept. A deal is not a row in a database. It is a living network that unfolds through founders, co-investors, experts, internal debate, doubt, reversal, and conviction. What matters is not only what happened but how the judgment formed, who believed what, on what evidence, and when. Most systems lose exactly that, because they treat deals as documents and workflows instead of as evolving records of belief.

And none of this was ever about automating judgment, quite the opposite. Venture is built on relationships, networks, and human depth, and any system that ignores that will fail. Originalis exists to amplify judgment, not replace it. Technology should do the work. People should practice the craft.

WHAT THE DATABASE KEEPSNORTHWIND ROBOTICS · SERIES A · INVESTONE ROWWHAT ACTUALLY HAPPENEDFIRST SIGNALFIRST MEETINGINTERNAL DEBATEDOUBTNEW EVIDENCEREVERSALCONVICTIONWHO BELIEVED WHAT, ON WHAT EVIDENCE, AND WHEN. THE RECORD THE ROW CANNOT HOLD.
FIG. 03 · A RECORD OF BELIEF
04 · WHAT WE BUILT

What we built

Venture did not need another tool. It needed a system of work. A system where sourcing is access, where analysis evolves instead of being regenerated. Where diligence is continuous, not episodic. Where conversations are first-class context. Where the network is shared intelligence. Where memory compounds, so the firm gets sharper every quarter it keeps deciding.

That is what Originalis is. It is not a CRM. It is not a robo-investor. It is the place where the work of a venture firm actually lives, from the first quiet signal on a founder to the memo, the decision, the board seat, and the next fund raised on the record you already keep.

05 · TASTE AT THE EDGES

Taste at the edges

Every firm we talk to asks the same question before any other: is this one size fits all. It is not. The instruments are the same across firms. The judgment inside them is yours. Ori learns the way your firm decides: what you asked, what you passed on and why, which references you trust, what a good deal looks like to you and to each partner individually. That learning never leaves your walls.

Two firms can run Ori on the same deal and arrive at different answers, because the taste at the edges is theirs. That is the point. Your advantage is your judgment, and Ori is built to learn it, not to average it away.

ONE DEAL, EVERY FIRM READING ITTHE AVERAGE ANSWERFIRM A · INVESTFIRM B · PASSYOUR ADVANTAGE IS YOUR JUDGMENT
FIG. 04 · TASTE AT THE EDGES
06 · WHO BUILT IT

Who built it

Originalis was built from inside the craft, by practitioners who are doing the work of venture and building the system we wish we had, people who have done this work end to end rather than studied it from a distance. That is the foundation, and it is the difference you feel in every surface.

Every product is AI native now.
Originalis is venture native.