Controlled Early Access now opening

The local-first GIS + AI workbench
for applied spatial modeling.

AIMU brings spatial data, classical ML, and deep learning into one desktop workflow — without forcing your research data into the cloud. We are selecting early users from labs, universities, and technical teams before the public release.

Request Early Access Explore Use Cases
Windows 10/11 · Local execution · Access reviewed before public release
Works with Rasters, shapefiles, and tabular datasets — side by side in one project Ships with Random Forest, SVM, CNN, LSTM, GRU, and Time-Series Transformers On your machine Runs offline on Windows · License-based · No cloud account, no usage meters
AIMU Software Interface Spatial layers Model training Evaluation & metrics
Runs on your hardware
No cloud account · No usage meters · Data stays on-premise
One project, end-to-end — import, preprocess, train, compare, evaluate — saved together.

Why local-first isn’t just a label

Your data stays on your hardware

Sensitive rasters, field surveys, and client datasets never leave the machine — no upload, no third-party storage, no review cycle with IT.

No cloud meter running

Training a CNN for eight hours costs you electricity, not GPU-hours. No per-seat cloud fees for your lab or your students.

Works where your data works

Air-gapped networks, field laptops, restricted university environments — AIMU runs wherever Windows runs, online or offline.

Built for real technical work

Four things AIMU does that fragmented stacks struggle with — integrated GIS + AI, offline execution, reproducible projects, and honest scope.

One workflow, start to finish

Prepare spatial data, engineer features, train ML and DL models, and review results in one desktop app — no gluing tools together.

GIS and AI, natively integrated

Multi-raster sampling, shapefile support, coordinate systems, and interactive mapping sit alongside Random Forest, SVM, CNN, LSTM, GRU, and Time-Series Transformers.

Honest about scope

AIMU is a workbench, not a black box. Preprocessing, training, and evaluation are inspectable; you choose every model parameter, and every run is saved with its config.

Experiments you can trust

Preprocessing, model configuration, and evaluation are saved with each project, so you can reproduce runs, compare models, and hand work off.

Who AIMU is built for

Applied researchers

Running spatial modeling for environment, earth observation, hydrology, agriculture, or geoscience — and wanting one tool instead of five.

Engineers & analysts

Building ML and DL models on geospatial or tabular data without handing datasets to a third-party cloud.

Labs & technical teams

Standardizing a local, reproducible AI pipeline across students and staff — without cloud lock-in or recurring per-seat cloud costs.

Your project, end-to-end

A typical AIMU project moves through five stages — load, preprocess, train, compare, evaluate — all saved in one reproducible project file. Example workflow: import rasters and shapefiles, prepare predictors, train a model, compare metrics, and save the full experiment in one project.

Load & explore data

Import rasters, shapefiles, and tabular datasets; browse coordinate systems and spatial layers.

Preprocess & engineer

Handle missing data, normalize, transform, and engineer features ready for spatial ML.

Train ML & DL models

Random Forest, SVM, and other classic ML, plus CNN, LSTM, GRU, and Time-Series Transformers.

Experiment & compare

Run and compare experiments side by side — with tuning, clear metrics, and reproducible project state so you can justify every model choice.

Evaluate & reuse

Inspect results, export outputs, and reload saved models to run predictions on new data.

How AIMU is built

Local-first

Your data stays on your hardware. No silent telemetry of project contents, no remote training queue, no vendor lock-in.

Desktop-native

Runs on your Windows machine, online or offline. Built to sit next to your data.

Reproducible

Projects capture preprocessing, model configuration, and evaluation together.

Controlled access first

Early users are reviewed before access. Pricing and public downloads reopen when the release funnel is ready.

Frequently Asked Questions

Quick answers to the things customers ask most.

No. AIMU runs spatial processing, training, and evaluation on your local hardware. Online access is used for controlled access and license-related checks, not for cloud training.

Yes. You can deactivate AIMU on one machine and reactivate it on another (e.g., upgrading your lab laptop).

No functional restrictions. The Student license includes the full workbench feature set but is legally restricted to non-commercial, academic use.

Yes, for Team/Lab licenses. Please contact our sales team to arrange invoice billing and university procurement.

Request controlled early access to AIMU.

AIMU is entering private beta with selected researchers, labs, and technical teams. Tell us your workflow and we’ll review whether the current build fits your use case.

Request Early Access Explore use cases