ruru.bio
Decision intelligence for living systems

Catch the loss
before it's visible.

Loss in a living system behaves nothing like a machine breaking down — a herd, a shoal, a culture, a crop, a microbial reactor. The damage starts out of sight, the threshold keeps shifting, and there is seldom a single right answer. ruru forecasts the hidden state, finds the optimal move within real limits, and shows the odds — in time to act.

Predict the risk · Optimise the response · Minimise the loss

Start with a retrospective on your data See how it works
We prove it on your history before you commit. Land, sea, and the biology AI is starting to design.
The hidden state moves firstsign lags harm
MOVING THRESHOLD LOSS BECOMES VISIBLE ACT NOW
Hidden riskWhat you can seeThe line that bites
Why we exist

The real fight is before you know it.

ruru exists for a specific kind of loss: where prevention costs a fraction of recovery, where most of the damage has no cure, and where by the time the loss is measurable it is already done.

Half the losses we work on never show on a gauge. So the data has to beat observation, the maths has to be right, and the recommended call has to be trusted enough that the right person acts on a threat they cannot see. Getting all three to hold is harder than any single forecast — and it's the whole job.

Who it's for

Find your system.

If you make decisions about a living system under uncertainty — or you carry the loss when it fails — one of these is you.

Growers & co-ops

Protect the season

Disease, heat, frost and blooms cost you before you can see them — and biologicals only work in a narrow window.

ruru gives you a daily watch-list: what's coming, the best action, and the confidence — early enough to act.
Aquaculture & fisheries

Read the water

Gill disease, oxygen crashes, thermal mortality and shifting stocks hit the pen or the quota before the signal is clear.

ruru forecasts the crash and the shoal, and optimises the response under stock, quota and capacity limits.
Biologicals companies

Make efficacy predictable

Your product works in the trial and fails in the field — because efficacy is context-dependent, and adoption stalls.

ruru forecasts where & when it will work, turning erratic field results into a targeted recommendation.
Biomanufacturers & therapy makers

Save the batch

Bioreactor drift, contamination and potency loss show up after the batch is lost — and some batches are a patient's only dose.

ruru forecasts batch trajectory early and optimises the feed / harvest / hold / abort call under scarce capacity.
AI-biotech & design labs

Close the reality gap

You can design biology faster than you can trust it in the real world. Design is cheap; knowing how it behaves once released is not.

ruru is the reliability & deployment layer downstream of your design engine — model-agnostic, governed, uncertainty-first.
Insurers, levy bodies & agencies

Own the loss earlier

You carry the aggregate loss across many operators, but you see it only after it lands — in claims, in yields, in a die-off.

ruru turns the forecast into the trigger and the early-warning — priced, governed, and uncertainty-explicit across a portfolio.
Our test

We don't chase every problem. We screen for one shape.

These are the six signs we look for. Few problems show all six, and we don't need them to. The more a problem shows, the more a decision layer moves the number rather than just dressing up a forecast.

01
The sign lags the harm
A hidden state moves first: the heat load, the liver dose, the infection. The visible signal only confirms it once the damage is underway — so a sensor or a forecast alone is already too late.
02
The threshold moves
The level where it bites shifts with humidity, crop stage, litter, contacts, the individual animal. It's a model, not a dial you can set.
03
The damage compounds
Delayed and cumulative, and often it can't be undone. A missed window doesn't stay small — it grows.
04
The optimal action changes
There are several ways to respond, and the optimal one depends on the conditions. Sometimes the optimal move is to do nothing.
05
The fix competes for the scarce resource
Cooling burns the water you're trying to save; zinc costs money and risks toxicity; crews, gear and cleanroom slots are finite. A real trade-off, not a free lever.
06
Someone clearly carries the loss
A risk owner with the mandate and the budget to act: an operator, a levy body, an agency, a processor, an insurer. Foresight without an owner changes nothing.

The more of these a problem shows, the more forecasting plus optimisation moves the number. Where the fit is partial, we say so.

How it works

Your data in. A decision, with the odds, out.

No rip-and-replace. We wrap your existing signals in a decision layer — and loop forecast and optimisation together, each improving the other.

01 · FORECAST

See the hidden state

Probabilistic, uncertainty-aware forecasting built for the sparse, noisy data you actually have — a few sensors, monthly tests, patchy trials, a district-blurred signal.

In: your sensor / test / batch / weather / trial data. Out: the optimal path and where reality is heading, with a confidence band.
OUR EDGE
02 · OPTIMISE

Choose the best move

Constraint-based optimisation turns the forecast into the best action under your real limits — crew, cooling, product, cleanroom slots — re-optimised as the event unfolds.

Out: a ranked action (do this, then this; sometimes: do nothing) with the expected loss avoided.
03 · COMMUNICATE

Act with confidence

Every recommendation arrives as evidence and a probability — never a bare number — so even a sceptical operator can act on a threat they can't yet see, and defend the call.

Out: an alert before the loss, and the odds behind it. Integrates with your dashboards / ops flow.

Forecast and optimisation are not a one-way pipeline. We loop between them, each improving the other, so the forecast is built to drive the decision. One shared engine across land, sea and introduced biology — the maths is the same; the fluency is specific.

Where it applies

Invisible, valuable, and treated as 'unsolvable'.

Different systems, one mathematical shape. The industry can already sense something — a THI reading, a gill score, a bloom map, a bioreactor trace. What none of it delivers is the optimal action, early enough, under real constraints.

Landdetails
Where we start · desert & Gulf dairy

Heat-stress dairy

Land · desert & Gulf dairy

Heat-stress dairy

In hot, humid climates a confined herd sits above the heat-stress line much of the year. About half the milk loss is metabolic, and shows on no gauge. Forecast it pen by pen, then optimise cooling under a hard water budget where every litre is desalinated.

~30% daily milk lost, summer vs winter*
Landdetails
Pasture · NZ dairy & sheep

Facial eczema

Land · pasture livestock

Facial eczema

Spore load rises and silent liver damage accumulates weeks before any visible symptom. Forecast the spore-and-dose trajectory paddock by paddock, then optimise zinc, spore-monitoring and grazing moves against cost and toxicity limits.

Sub-clinical damage before the first sign
Landdetails
Orchard & vine · bloom risk

Frost & spray windows

Land · horticulture

Frost & spray windows

Latent infection and frost injury are decided at bloom, hidden for weeks, and the kill-threshold keeps moving. Forecast block-level risk, then optimise which rows get scarce wind machines, heat or a spray first — under PHI and spray-count limits.

Spatial risk, machine & spray allocation
Seadetails
Salmon aquaculture · gill health

Gill disease onset

Sea · finfish aquaculture

Gill disease onset

Amoebic and complex gill disease thickens tissue and degrades respiration well before fish show it. Forecast onset probability from sparse gill scores plus temperature, then schedule scarce freshwater-bath capacity before mortality climbs.

Onset forecast, bath-capacity timing
Seadetails
Fisheries · migrating stocks

The moving shoal

Sea · wild fisheries

The moving shoal

Stocks redistribute with a warming, shifting ocean, and the fishing-ground call trades catch against bycatch, quota and fuel. Forecast stock distribution, then optimise allocation under quota and access limits, trip by trip.

Distribution forecast, quota-aware routing
Waterdetails
Water utilities · algal blooms

Bloom & toxin response

Water · catchments & intakes

Bloom & toxin response

Cyanotoxin concentrates near a raw-water intake hours-to-days before routine sampling confirms an exceedance. Forecast breakthrough with calibrated lead time, then trigger pre-emptive treatment or intake switching before the limit is crossed.

Breakthrough forecast, pre-emptive dosing
AI-biodetails
Ag-biologicals · field efficacy

The reality gap

Introduced biology · biologicals

The reality gap

A biological works in the trial and fails erratically in the field because efficacy is context-dependent. Forecast where and when it will actually work, then optimise application timing and placement — turning erratic results into a defensible recommendation.

Context-to-performance forecast
AI-biodetails
Biomanufacturing · cell & gene

Save the batch

Introduced biology · biomanufacturing

Save the batch

Bioreactor drift, contamination and potency loss surface after the batch is lost — and a cell-or-gene batch can be a patient's only dose. Forecast batch trajectory early, then optimise the feed / harvest / hold / abort call under scarce cleanroom capacity.

Early trajectory, hold/abort optimisation

*Loss figures shown are industry or published ranges for the problem, not ruru results. We never put a saved-loss number on a slide that your own data hasn't earned — that's what the retrospective below is for. Photography is illustrative.

Start here — no claims you have to take on faith

We prove it on your own history first.

Give us a season, a run of batches, or a set of past trials. We'll replay your own history with our optimiser in the loop — same budget, same teams, a different allocation — and show you, decision by decision, where ruru would have flagged the loss, how early, and what the best action would have been. If it wouldn't have helped, you'll see that too.

Book a free retrospectiveTalk about a pilot
  • Runs on data you already have
  • A back-test on your history — early-warning lead time, quantified
  • Honest result: we show where it helps and where it doesn't
  • You keep the analysis whether or not we go further
  • Your data stays yours — rights and use agreed up front
Two ways in · one engine

Read the system. Estimate the interruption.

Read the system

Forecast a natural system

See the hidden state of a living system and act before the loss is visible — disease, heat, blooms, stocks. This is the applied work ruru already runs, extended across land and sea.

Estimate the interruption

Predict an introduced agent

Before a new biological enters a living system — engineered, AI-designed, or not — forecast how it will actually behave: where, when, whether, and what could go irreversibly wrong.

One model of the living system, two questions: P(future | observe) to read it, P(future | do(introduce)) to estimate the interruption.

The team

Built for hard decision problems.

Our scientists sit behind the forecasting, optimisation and uncertainty methods that others build on — a combined 50,000+ citations on Google Scholar. We bring in domain specialists as a problem demands, led by a CEO who turns advanced science into tools teams trust and use.

Dr Christoph Bergmeir
Core · Forecasting
Dr Christoph Bergmeir
Probabilistic & adaptive forecasting. Co-author of NeuralProphet; Clarivate top-1% forecasting papers.
Google Scholar ↗
Prof. Peter Stuckey
Core · Optimisation
Prof. Peter Stuckey
Constraint programming & optimisation. Creator of MiniZinc. AAAI Fellow · Google Eureka Prize.
Google Scholar ↗
Prof. Wray Buntine
Core · Uncertainty
Prof. Wray Buntine
Bayesian machine learning & uncertainty. Ex-NASA Ames, Berkeley, Google.
Google Scholar ↗
Dr Abishek Sriramulu
Core · Contextual AI
Dr Abishek Sriramulu
Graph, spatio-temporal & multimodal AI; adaptive dependency-learning GNNs.
Google Scholar ↗
Dr Frits de Nijs
Core · Optimisation
Dr Frits de Nijs
Risk-aware multi-agent & stochastic optimisation, reinforcement learning. 3rd, NeurIPS 2021 ML4CO.
Google Scholar ↗
Paul Shale
CEO · Product & Strategy
Paul Shale
Turns advanced science into tools teams trust and use. Law, finance, Harvard (disruption); startups across NZ, AU and the USA.

Domain experts set the constraints; the team builds the maths. They prove each method in the open literature first.

50,000+
citations across the team's work (Google Scholar)
1,000+
peer-reviewed publications, combined
Top 1%
Clarivate highly-cited in forecasting
The research

We prove it before we promise it.

The optimisation-under-uncertainty at ruru's core is published work, not a pitch. Our team co-authored the Predict+Optimize benchmark for renewable-energy scheduling — the same forecast-then-optimise shape ruru applies to living-system risk, which is why the method carries across problems.

Peer-reviewed · verified
Predict+Optimize Problem in Renewable Energy Scheduling ↗
IEEE Access, vol. 13, pp. 60064–60087, 2025
DOI 10.1109/ACCESS.2025.3555393
Four of ruru's core scientists are among the authors. A competition benchmark in energy scheduling, not agriculture.

Prove, not promise.

Before we ask anyone to build on faith, we replay their own history with our optimiser in the loop — same budget, same teams, a different allocation — and show, decision by decision, what it would have saved against what actually happened.

If the number holds, there's a position worth funding. If it doesn't, we recalibrate before we build. We never put a saved-loss figure on a slide that real data hasn't earned.

Selected publications · the methods behind ruru
  • Predict+Optimize Problem in Renewable Energy SchedulingBergmeir, de Nijs, Genov, Sriramulu, Abolghasemi, Bean et al · IEEE Access, 2025 · DOI ↗
  • Local and global trend Bayesian exponential smoothing modelsSmyl, Bergmeir, Dokumentov, Long, Wibowo, Schmidt · International Journal of Forecasting, 2025
  • MSTL: seasonal-trend decomposition for time series with multiple seasonal patternsBandara, Hyndman, Bergmeir · International Journal of Operational Research, 2025
  • Online guidance graph optimization for lifelong multi-agent path findingZang, Zhang, Harabor, Stuckey, Li · AAAI, 2025

Each scientist's full publication record is on their Google Scholar, linked in the team section above.

The vision path

A route into the imminent agri future.

As AI makes designing biology cheap, the bottleneck becomes trust, test-selection and control — forecasting-and-optimisation problems, not generation problems. We start where the loss is immediate today, and the same engine carries forward.

Today

Read natural systems

The applied book across land and sea. Revenue, and a growing corpus of state→outcome data.

Now emerging

Biologicals reliability

Forecast where and when a biological actually works — the gap throttling adoption.

Next

Biomanufacturing & therapy

Batch-reliability for bioreactors and one-shot cell & gene therapies.

Horizon

AI-derived biology

The reliability, deployment & containment layer for organisms AI designs.

How we operate

We show the odds.

01

Uncertainty on the face of it

Every output is a probability with an interval, not a confident point. If we aren't sure, you see it.

02

Your data stays yours

Rights and use agreed before we start. The value we build on your data is yours.

03

Stewardship by design

The maths that optimises benefit could, sign-flipped, optimise harm. Introduced-agent work is access-controlled and dual-use aware.

Start here

Bring us a decision you're guessing at.

A herd, an orchard, a catchment, a shoal, a batch, a release. We'll show you the hidden state, the best move, and the odds — on your own data, first.

Book a free retrospectiveSee where it applies
Investor or research partner? Talk to us about the platform →