SGNLIn development

SGNL: Your next 3–5 priorities.
Ready to act on.

An AI to-do list that picks your most impactful work for the next 18 waking hours, then prepares it.

Built in-house. The ring and first handoff loop are fixture-tested.

The work around your priorities03 / 05
How SGNL turns incoming work into a founder's ring Product design: email, meetings and code give context for your goals. SGNL prepares work around three to five human priorities. EmailMeetingsCode Your goalsVerify DraftsDemosTests + video Next 18 hours3–5prioritiesOutcomes update the work
SGNL does the work around your work.

Draft the email. Load the demo data. Run the smoke tests.

Product concept, not live data. The operating loop is in development.

Arrive with the
work prepared.

The 3 to 5 SGNLs are your priorities. The system prepares the rest.

  • The email is draftedYou review it and approve sending.
  • The demo is readyTailored for an investor meeting, data pack loaded ahead of time.
  • The feature is testedPlaywright smoke tests in Chrome, recorded for the CEO to review.

These describe the assistance layer being developed.

SGNL runs the tests.
You review the evidence.

  1. 01A feature ships

    Deployment triggers verification automatically.

  2. 02SGNL plans the checks

    A browser test plan: what changed and what should happen.

  3. 03AI runs the browser

    Playwright drives Chrome and checks results against expected behavior.

  4. 04The run is recorded

    Recordings and results show what failed, and where.

  5. 05The CEO reviews

    Watch the recordings and make the product call.

Target workflow, in development. You review the evidence and decide.

AI engineering

Turn context into prepared work.

Structured generation prepares the work; rules keep the ring small.

Engineering detail: AI engineering

Components

  • Python
  • Pydantic
  • OpenAI-compatible gateway
  • Jev shadow evaluation

What is in the system

  • Rules rank work against the owner’s objectives.
  • Jev runs the same judgment in shadow mode, without changing a live decision.
  • Model outputs are structured and validated.
  • Evidence is redacted, versioned and treated as data, never instructions.

Systems engineering

Keep the right work current and moving.

One system delivers the ring, carries work across tools and checks fresh evidence.

Engineering detail: systems engineering

Stack

  • Next.js
  • Python
  • Supabase
  • Postgres
  • GitHub
  • Linear

What keeps it reliable

  • Next.js on Vercel shows the ring on an 18-hour cycle, refreshed every 20 seconds.
  • A Python worker on Railway processes events in Supabase Postgres.
  • Progress and Blocked actions, a five-second undo and invite-only sign-in keep people in control.
  • Rechecks, a last-checked time and worker heartbeats expose stale claims and failed handoffs.
  • Time-limited claims and bounded retries keep interrupted work with its packet.
  • Fixtures exercise the web and worker paths. Production validation is outstanding.

The same pattern we build for clients.

Connect the tools, find one expensive handoff, let AI prepare the work.

  • Evidence before recommendationEvery signal shows why it exists.
  • Safe work before interruptionThe system prepares what it can first.
  • Human control at the boundaryApprovals, money movement and consequential changes stay explicit.

What should AI take off your plate?

Bring one repeated workflow. We will say what can be automated safely and what a first release should cost.

Book a call