Hosted prediction service

From your data to live decisions in a day

Bring labelled data and call an API. MathFi.ai refines your data, builds a production-grade model, validates it, and hosts and serves it, the same day. No ML infrastructure to run, no data-science team required, and every decision is deterministic: same data in, same answer out. It is at its best on the decisions that cost you most, the rare ones your current models miss.

Rare-class edge
Catches more of the rare, high-cost cases (the fraud, the default, the failure) that standard models miss
Same day
From labelled data to a live production model, typically within one working day
Deterministic
Same data in, same decision out. Every prediction can be reproduced and audited
DS optional
If you have a data-science team, they go faster. If not, you still go faster

Your data already knows. Getting it to tell you is the problem.

The answers are in your data.

Which claims are fraudulent. Which customers are about to leave. Which applicants will repay. Years of labelled history sit in your systems, and the patterns are in there.

Turning data into decisions has been slow and scarce.

It has meant specialists you cannot hire fast enough and a quarter of work per model: cleaning, feature engineering, tuning, validation, deployment, monitoring. Most organisations get a handful of models a year through that pipeline, if any.

So most of the value never ships.

The predictable fraud gets paid out. The savable customers leave. The good applicants get declined. Not because the signal is missing, but because the route from data to production has been too long to walk more than a few times a year. That is the bottleneck MathFi.ai removes.

Fraud, churn, credit, claims: the decisions you cannot afford to get wrong

MathFi.ai works on any decision that lives in labelled tabular data: risk, retention, operations, quality, safety. But the fastest payback is where the case you care about is rare and each miss is expensive. That is where standard models are weakest and MathFi.ai is strongest, so it is where most teams start.

Fraud detection

Surface more fraudulent transactions, claims and applications for review, with fewer honest customers flagged along the way.

Customer churn

Identify the accounts likely to leave while there is still time to keep them, and aim retention spend where it changes the outcome.

Credit and default risk

Approve more of the customers who will repay and catch more of those who will not, instead of trading one error for the other.

Insurance claims

Pay honest claims fast and route the small minority that deserve investigation.

Equipment and operational failure

Predict rare breakdowns from operational data before they become downtime.

Clinical and safety risk

Flag the rare, high-stakes cases that population-level screening misses.

Where teams start: rare, high-cost decisions. Fraud, churn, credit and default risk, and claims are the entry wedge, because that is where a same-day model pays for itself first, and where our benchmark lead is largest.

Morning: your data. Afternoon: a served model.

The flow below is not a roadmap; it is a working day. You provide labelled data in the morning. Feature Refinery cleans it and selects the features that matter. Model Crucible builds the model and validates it on held-back data. MathFi.ai publishes and serves it. By the end of the day your application, or your agent, is calling a live endpoint and getting deterministic decisions back.

You / Your Agent
Choose the decision, bring the data
Pick what to predict and supply a wide set of raw features, with labelled outcomes.
MathFi.ai
Feature Refinery
Cleans the raw, noisy data and selects the features that matter.
MathFi.ai
Model Crucible
Builds the model and validates it on held-back data.
MathFi.ai
Publish
The validated model goes live as a hosted endpoint on our systems.
You / Your Agent
Production inference
Your app or agent calls the model through an API and gets a decision back.
The whole flow can complete in a single day.

A day is possible because of what is underneath. MathFi.ai runs on Cellular Balanced Learning (CBL), a genuinely novel machine-learning architecture our co-founders invented after decades of building production models the slow way. CBL is not a wrapper around open-source libraries; it changes how the model is built. Rather than searching across more than a hundred settings for a good decision boundary, CBL constructs dynamic cell boundaries that adapt to the structure of your data as it trains, and the relationships between them are learned inside the architecture.

That is why a CBL model exposes two hyperparameters where a gradient-boosted model exposes more than a hundred. The tuning decisions a specialist normally makes by hand, learning rate, depth, sampling, regularisation and the rest, are internalised in how CBL is built. The expertise sits in the architecture, not the operator, which is what lets a team without deep ML staff get a strong model, and lets a team that has one skip the tuning marathon.

It also earns its keep on quality. Most business prediction problems suffer from class imbalance: the cases that matter are a small minority of the data, and conventionally trained models sacrifice them for overall accuracy. Because CBL handles imbalance structurally rather than through oversampling patches, the same-day model is also the one with the strongest minority-class recall and precision in our benchmark. And it is a service, not a software project: MathFi.ai hosts and serves the model, with every decision reproducible and auditable.

Read how CBL works in the docs →

The rare, costly decisions are where MathFi.ai pulls ahead

Most teams have tried at least one route to better predictions, and two things go wrong. The hard models, defeated by feature selection, more than a hundred hyperparameters and overfitting, often never reach production, so the value never lands. The ones that do ship are weakest exactly on the rare, costly cases. That gap, felt most by smaller teams frustrated at how little their current approach returns, is what MathFi.ai is built to close.

Versus building it yourself

Doing this in-house means months of work per model and specialists who are hard to hire and harder to keep. Picking the right features, managing more than a hundred hyperparameters and fighting overfitting means many attempts never reach production at all, and the ones that do patch rare events rather than solve them, with oversampling tricks like SMOTE and class weights. MathFi.ai handles the rare cases structurally, in how the model itself is built.

Versus heavyweight enterprise AutoML

The large AutoML platforms are expensive and complex to adopt, and at the end of the tuning run they hand you the same gradient-boosted models your team already runs, with the same blind spot on rare events. You pay platform prices for automation of the status quo.

Versus newer tabular and foundation-model AI

The new generation of tabular foundation models is genuinely interesting, but it is not deterministic. If you cannot reproduce a decision, you cannot audit it, explain it to a regulator, or debug it when it goes wrong. MathFi.ai gives the same decision on the same data, every time.

Versus general-purpose LLMs

LLMs are probabilistic by design. That is the right property for language and the wrong one for a decision that has to be trusted: a lending call, a fraud flag, a churn intervention that must be right and repeatable. Use an LLM to talk. Use MathFi.ai to decide.

Not one model, the best of many

The right model for your data, automatically

MathFi.ai runs several competing algorithms and automatically selects the best-performing model for your data, so you are not betting on a single approach, and you are not hand-tuning your way to it.

Fewer missed cases and fewer false alarms at the same time; deterministic and auditable; no data-science team required; and it reaches production the same day.

Measured where it matters: the rare, costly cases

Under class imbalance, headline accuracy hides failure, so we measure the minority class: the fraud, the default, the failure, the case you actually care about. Here is how MathFi.ai performs against the five libraries most teams already run, each on an identical train and test split.

Minority-class F1 score, in percent (higher is better). F1 balances how many of the rare cases you catch against how many false alarms you raise. MathFi.ai versus the best of five standard libraries (XGBoost, LightGBM, CatBoost, Random Forest, Logistic Regression) on each public dataset.
DecisionMathFi.ai · F1 %Best other · F1 %
Credit approval88.5086.21
Consumer default risk63.6059.70
Signal integrity / anomaly detection97.4491.89
Cardiac risk detection84.2082.63
Stroke screening27.0420.31

Highest minority-class F1 in 6 of 7 datasets, and the highest or joint-highest recall in all 7.

96.24%

Of rare warranty fraud surfaced for review, against 82 to 84% for the gradient-boosting methods. Every point in that gap is fraud that would otherwise have been paid out.

+2.29 pts

Higher F1 than a fully tuned XGBoost on credit approval, with 4.24 points more precision at the same recall: fewer good customers wrongly declined.

Accuracy is a dangerous metric for rare-class events. On a public stroke-screening dataset, one competitor scored about 90% accuracy yet found only around 26% of the patients who went on to have a stroke: it earned its score by predicting "no stroke" almost every time. MathFi.ai found 66% of the patients who went on to have a stroke. When the event you care about is rare, accuracy rewards the model that ignores it; recall and precision on the rare class tell you whether the model actually works.

MathFi.ai benchmark, July 2026. Seven public datasets. Identical train and test splits. Five standard libraries as baselines.

Where a method beats us on a secondary metric, our study says so plainly.

See the full benchmark study, all seven datasets →

Which of these sounds like you?

Mid-market, little or no ML staff

You know the answers are in your data. Fraud losses, churn, bad debt: you can see the cost lines, but building models has meant hiring a team you cannot find or afford. MathFi.ai gives you a production prediction capability without the department. Bring labelled data; get a served model.

Data-science lead with a backlog

You have the team, and a queue of rare-event problems that resist the standard toolkit. MathFi.ai clears that queue: same-day models with minority-class recall your current stack cannot reach, deterministic outputs your risk committee will accept, and your scientists freed for the work only they can do.

Regulated and agentic builders

You operate where every decision may need to be explained, to a regulator, an auditor or a customer. Or you are building agents that act on predictions and cannot ship on a model that answers differently on the same input. MathFi.ai is deterministic end to end, so every decision is reproducible on demand.

Agent-ready today

MathFi.ai models are callable from your agents, copilots and applications through a documented REST API with an OpenAPI spec. Every prediction returns a deterministic, auditable decision your systems can log, replay and explain from the exact inputs the model saw. This is the difference between an agent you demo and an agent you deploy. Native MCP support is on the roadmap; today MathFi.ai is agent-usable through standard API integration.

Read the API docs →

Built by people who lived this problem for decades

MathFi.ai was built by three operators who spent their careers shipping production machine learning and data systems at scale, and hit the exact wall that Cellular Balanced Learning was invented to remove. Between them, more than 20 granted patents.

SA
CEO & Co-Founder

Inventor of Cellular Balanced Learning. 25 years building machine learning, algorithms and patents at Fujitsu Laboratories, Tesco, import.io and Syncron. PhD from the University of Surrey.

LinkedIn
DF
CTO & Co-Founder

17 years in backend and platform engineering, including at Tesco. Built the platform that takes CBL from research to production scale.

LinkedIn
SO
Non-Exec Chair & Co-Founder

25 years leading product and technology at scale across Amazon, Tesco, GfK and startups.

LinkedIn

Bring us the prediction you most need to get right.

We are choosing a small number of design partners. Bring a decision your current models get wrong too often, or a prediction you have never been able to build at all. We build a high-performance model on your own data and show you the lift against your own held-back validation data. This is for leaders who want to get past the tooling and straight to impact.

GenAI taught machines to talk. The next decade is teaching machines and agents to decide, and those decisions must be right, repeatable and auditable. MathFi.ai is the deterministic layer that GenAI calls when a decision must hold up: the prediction utility beneath the agentic enterprise.