Meter-to-grade infrastructure

Evidence cheap enough for a QR code.

Manual verification cannot sit inside a low-margin consumer product. Autograde connects industrial telemetry to a deterministic evidence pipeline, producing batch records at ¥2–5 each — with no human issuing every label.

Pipeline

Telemetry in. A bounded evidence record out.

Automation changes production cost, not the rules. Unsupported records still abstain.

01 · CONNECT

PLC / SCADA input

Energy, flow, runtime, output and calibration events arrive through signed gateway messages.

02 · NORMALISE

Canonical telemetry

Clock, units, missing intervals and meter identity are normalised before any footprint is computed.

03 · GRADE

Deterministic rules

Boundary, factors, allocation and metred share execute under one versioned method.

04 · PUBLISH

Commit and serve QR

SHA-256 commitment, public anchor, revocation slot and a product-ready verification URL.

POST /v2/autograde/batches { "site_id": "SITE-042", "meter_stream": "sha256:…", "batch_output": 480, "functional_unit": "plantlet", "method": "CJV-MTH-2.0" } → grade: "A" | "B" | "C" | "ABSTAIN" → unit_id, hash, anchor_status, verification_url
Economics

Automation is the only route to unit-price evidence

A fast-moving consumer product cannot carry ¥40–80 of evidence cost per small batch. Autograde is designed for high-volume, stable production environments where telemetry already exists.

Automated batch record¥2–5 / batch
SKU / annual volume planQuoted by volume
Gateway integrationOne-off scoped fee
Public QR verificationIncluded
Human technical reviewSampled / exception-based
ABSTENTION IS AN OUTPUT

Automation does not mean automatic approval

Missing intervals, stale calibration, inconsistent output or an unapproved factor source produces ABSTAIN. No label is issued. The exception enters the review queue with the exact failed rule.

Fit

Where Autograde belongs

Stable, instrumented production. Not a substitute for measuring an uninstrumented supply chain.

Good fit
  • Controlled-environment agriculture and nurseries
  • Continuous food, beverage and industrial production
  • High-volume SKUs with repeat boundaries
  • Retailers requiring QR-accessible product evidence
Not a fit
  • Invoice-only energy data
  • Unstable product boundaries or untracked batch mixing
  • Uncalibrated meters and unverifiable clock histories
  • A request to convert self-declaration into certification
Assess telemetry readiness → Read the API model