voice-d-founder

Voice D: founder-first-person — we built this because we needed it

Metadata

Statusdone
Assignedagent-251
Agent identity8da3b6fc81685ca44a4e15eb35307ab90ce3f0500e9c08b8b0caae848f7d4ce9
Created2026-05-02T02:42:53.612981717+00:00
Started2026-05-02T02:45:02.785909076+00:00
Completed2026-05-02T02:47:48.360105651+00:00
Tagsgrant,urgent,website,voice-study, eval-scheduled
Eval score0.95
└ blocking impact1.00
└ completeness1.00
└ constraint fidelity0.85
└ coordination overhead1.00
└ correctness0.95
└ downstream usability0.98
└ efficiency0.95
└ intent fidelity0.92
└ style adherence0.98

Description

Voice D: founder first-person

Drop into 'we' voice. Why we built this. What we run on it. What we still don't know. The personal voice answers Erik's 'too self-aware' diagnosis differently — text that's self-aware AS THE FOUNDERS rather than self-aware AS THE TEXT is fine. Honest. Direct. People are talking, not abstractions describing themselves.

Rough paragraph to replace

AI agents can now search literature, write code, analyze data, and design molecules. But the durable problem is organizational: how people and machines coordinate across days or weeks, preserve judgment, expose evidence, and stay responsive to participants.

We build open-source products, live demonstrations, and organizational methods that make hybrid human-AI work auditable, reproducible, and inspectable. Our first product is WorkGraph: a Rust task-coordination system where humans and AI agents work in the same dependency graph, with claims, execution logs, handoffs, completions, artifacts, and histories preserved for review.

WorkGraph is the instrument, not the whole institution. It gives Poietic a proof surface for the theory of organizational patterns: product, demos, research practice, and organizational design should all make human and machine collaboration legible and responsive to its participants. That public-benefit purpose is the operating constraint: the tools should expose the work, the handoffs, the failures, and the evidence trail.

Shared context

Erik diagnosis

"too self-aware. lol. can you run a team on it? it's kind of uncanny."

The current text describes itself meta-philosophically ("WorkGraph is the instrument, not the whole institution. It gives Poietic a proof surface for the theory of organizational patterns"). It is text talking ABOUT itself rather than text DOING the work of explaining the company. A reader asking "can you run a team on it?" should be able to look at the page and answer YES, here is how. Not get a paragraph about proof surfaces.

Constraints all voices share

  • PBC public benefit statement is the foundation: "to make human and machine collaboration legible and responsive to its participants" (Delaware filing, verbatim).
  • Founder order Erik / Luca / Vaughn whenever named.
  • Erik dogfooding instruction: "they build this because they want to USE it. Their track records prove they know scientific organizations and how to run real resource projects."
  • No em-dashes (style rule).
  • No PI / lead PI language.
  • No "in production today" or "we wrote this proposal with WorkGraph" recursion overclaim — instead: WorkGraph is in active use across founders' research and operations; the incorporation trace is publicly browsable at https://ulivo.poietic.life/wg/feeds/incorporation-trace/
  • Length: roughly the same as the rough paragraph (3 paragraphs, ~200-250 words). Can compress further if it works.
  • Match the existing dark monospace landing page aesthetic (no flowery prose, no marketing fluff).

Output

A single drop-in replacement section. Plus 50-word self-justification ("here is the voice I am writing in and why it answers Erik's diagnosis").

Output

~/poietic.life/notes/website-voice-d-firstperson.md containing:

  • The replacement text (3 paragraphs, drop-in)
  • 50-word self-justification

Depends on

Required by

Log