Skip to content
——:——:——Zchrononav-v2.0.4 / refinery-r1EVIDENCE BEFORE CLAIM

EDEN OS · VALUE-AWARE CONTROL LAYER

Use less.Prove more.

EDEN is a value-aware control layer for computation. It determines what work matters, what can be reused or removed, when remaining work should execute, and how those decisions can be independently verified.

  • KEEP

    useful work

  • VOID

    avoidable work

  • VERIFY

    evidence bound

  1. 01SENSE
  2. 02DECOMPOSE
  3. 03MEASURE
  4. 04KEEP / VOID
  5. 05REGENERATE
  6. 06NAVIGATE
  7. 07VERIFY

Operating system

Four modules. One control architecture.

Evidence ladder

Don’t trust the claim. Verify the evidence.

EDEN distinguishes evidence quality instead of presenting all output as fact. Anything simulated says so. Anything demonstrated only locally says so. AOK enforces that sentence before a capability may execute or speak in public.

  1. 01

    IMPLEMENTED

    The code path exists and ran.

  2. 02

    MEASURED

    A physical or instrumented quantity was bound.

  3. 03

    REPRODUCIBLE

    An independent run can repeat the result.

  4. 04

    MODELLED

    Computed geometry or policy — not a meter.

  5. 05

    PROPOSED

    A design or intent, not an observation.

SHA

Marbles

A cryptographic evidence container. Not a token.

A Marble binds provenance, policy, commitments and an evidence class to a hashed identity. It can describe a bounded quantity, an event, a resource, an outcome or a state transition. Plaintext stays out of the record.

IMPLEMENTEDDEMONSTRATION

PROVENANCE · POLICY · MEASUREMENT LABEL · VERIFY OR VOID

OPEN MARBLES →

Investor-grade proof

What is proven, and under which label.

Every metric carries an evidence class. Simulated or modelled results are never presented as physical measurements.

CHRONONAV V2

REPRODUCIBLESIMULATED

1,847.21

10,000-trial frozen-policy blind validation

Mean true utility vs. 1,315.77 EDF

Simulated blind trials under a frozen policy. Not a production cluster measurement.

REUSE BENCHMARK

REPRODUCIBLESYNTHETIC

79.9%

High-reuse synthetic workload

Reuse at 50k workload size

Synthetic reference test. Real-world reuse is workload-specific and must be measured per deployment.

MERKLE VERIFICATION

REPRODUCIBLE

100 / 100

1 GiB commitment experiment

1,048,576 packets · sampled proofs valid

Offline reproducible experiment. Not an externally anchored public timestamp.

PHYSICAL EDGE NODE

IMPLEMENTEDPARTIALLY MEASURED

IMPLEMENTED

Android / Termux execution environment

Local compute, network and energy telemetry tested

Components ran on-device. This is not a claim of fleet-scale edge production.

Platform

Six modules. One control architecture.

AURA — DISTRIBUTED INFRASTRUCTURE R&D

Shown last, and labelled. Ambition without a truth panel is a pitch deck. This is the truth panel.

  • Orbital geometryMODELLED
  • SGP4 positioningMODELLED
  • DopplerMODELLED
  • FSPLMODELLED
  • Physical satellite RF receptionNOT MEASURED
  • Operational satellite networkNOT CLAIMED

Markets

One control architecture. Multiple constrained-compute markets.

These are not seven unrelated businesses. They are the same question asked wherever compute, energy, bandwidth or time is scarce.

  1. 01

    AI INFRASTRUCTURE

    Token, KV-cache and model-call waste under latency.

  2. 02

    DATA CENTRES

    What to run, defer, or drop when power and cooling bind.

  3. 03

    EDGE COMPUTE

    Useful work on devices that cannot pretend they are the cloud.

  4. 04

    TELECOMMUNICATIONS

    Prioritised payload under bandwidth and energy.

  5. 05

    RESILIENT / DEFENCE SYSTEMS

    Degraded networks still need a value order.

  6. 06

    FINANCIAL COMPUTE

    Deterministic replay, provenance, deadline-true scheduling.

  7. 07

    SPACE INFRASTRUCTURE

    Research path (AURA). Modelled, not operational.

Shadow mode

Prove the economics before changing production.

Shadow Mode is read-only observation. EDEN classifies a workload and computes a labelled baseline, counterfactual and delta. AOK holds production execute until a measured validation exists.

BASELINE

COUNTERFACTUAL

DELTA