Proof

Proof, not promises

Every Harvora surface ships with a measured accuracy stat on real New Zealand data — benchmarked July 2026, provenance documented, caveats stated up front. This page is our technical brief, in the open.

Impact + accuracy, per surface
Frost
94%
of frost nights caught
Why it matters

The average NZ growing location faces 23.4 damaging frost nights a season, with exposure running to tens of thousands of dollars per hectare.

Measured

Catches 94% of frost nights at the economically-correct alarm setting. Production overnight-low forecasts run 15–18% sharper than the public forecast (1.38 °C RMSE at 24 h). Climatology-style tools miss 66% of frost nights.

Annual-crop intelligence

Potato and open-field tomato, modelled at full platform depth

Crop-cycle context now flows through phenology, irrigation, disease, harvest readiness and economics. Four guarded blight models and three insect lifecycle stacks extend the same probabilistic serving contract across all 9 supported crop families and 13 NZ regions.

Late blight
7 days
probabilistic crop-cycle outlook
Why it matters

Potato and open-field tomato now carry crop-cycle-aware late-blight intelligence through the same national prediction, finance and UI paths as the established crop families.

Measured

A 1,000-sample surface blending 60% formula with a 40% guarded classifier when the packaged artifact schema and hash match. The engineering corpus contains 6,510 generated feature rows across the four annual blight models.

Hazard & summer-risk pack

Seven more surfaces, covering the risks summer brings

Hail, sunburn, splitting rain, fire weather, crew heat safety, seasonal dryness and field access — the perils that arrive with heat, ripe fruit and harvest machinery. Each one beats the raw forecast or climatology on held-out data, and each card states exactly what was measured.

Hail
~25×
better than chance at flagging hail days
Why it matters

Hail can wipe out a season's fruit in minutes — and no forecast anywhere can call it block by block. Knowing when the atmosphere is primed is the edge that matters.

Measured

Machine-learned models read the convective environment (CAPE, lifted index) up to 7 days out, beating the raw forecast at every lead across a 58,526-row corpus spanning 13 regions. On the days it flags, real convective events occur ~25× more often than chance — with fully calibrated probabilities (Brier 0.0095, ECE 0.008).

The engine underneath
1,000
Monte Carlo samples per prediction
~1.5 ms
Per full simulation
Bit-identical
Same seed, same result — auditable
OOD-guarded
Refuses to extrapolate beyond NZ
1.075 °C
Forecast meta-model MAE · 13 regions
24,217
Frost nights benchmarked · 20 NZ sites
2,103
Python tests passing · 87 optional skips
486
Web tests passing
35
Packaged ML artifacts
38
Versioned metadata records
×18
Surfaces green on the calibration gate
9 · 13
Crops · NZ regions modelled at equal depth
What we don't claim (yet)

Numbers you can trust exist only where a company says what it can't claim

Everything above is measured against real, independent data. A few things aren't provable yet — and rather than blur the line, we publish it. This discipline is the product.