AdaptivecityDiscuss your project

Independent AI product studio

Intelligence, grounded.

We design and build AI products for the physical world: sensing, spatial intelligence, and software people can act on.

Discuss your projectExplore the work
CANDIDATESURFACECONTINUOUSSAMPLEDDISCRETE
Sampling grid on drawn terrain

A signal becomes intelligence when it meets context, evidence, and a decision someone has to make.

  1. Sense

    Measure the world as it actually is.

  2. Understand

    Calibrate the reading and bound its uncertainty.

  3. Decide

    Rank what a person should look at first.

  4. Learn

    Return the outcome to the model that ranked it.

Illustrative practice model, not a claim about any one project
INLETPUMPCELLFIELD READINGSENSORREFERENCEFITTED MODEL
Illustrative sensing and calibration schematic

Research that meets the field.

We contributed to published research on portable methane sensing and machine-learning calibration, tested in laboratory and landfill settings.

Co-authored by Andrea Massara · Sensors, 2026

Measuring Methane Emissions in Ambient Air with a Low-Cost, Portable Sensor System: Focus on Scalability and Transferability of the Model.

Bertin, L.; Mentasti, M.; Pittorino, F.; Villa, V.; Zanni, E.; Viscardi, G.; Ponzani, Y.; Massara, A.; Roveri, M.; Dellaca’, R.; Capelli, L. Sensors 2026, 26(13), 4321. Published 7 July 2026.

Andrea Massara is credited for software and resources. The paper reports work within ESCAPE through Integraciones Digitales Gold SL (Indigo). It does not validate Landfill OS satellite screening.

Read the paper

Selected work

From observation to action.

Two scales of one practice. One reads a territory from orbit. The other reads a street from your pocket.

Landfill OSIn developmentTerritory scale

See what changed. Know where to look next.

Landfill OS links satellite observations to site context and evidence, so a team can tell which change is worth a site visit.

Illustrative screening sequence

Illustrative three-step diagram on synthetic terrain: two observation dates are compared cell by cell, site boundaries and evidence markers narrow the changed cells, and one cell is ranked as the candidate for field investigation.

Read the case notes

What this case demonstrates

Methane Mapper is the starting point: a project connecting field measurement to operational intelligence. Landfill OS extends that inquiry to satellite observations and evidence-led screening.

The illustrated sequence shows the intended decision flow. Satellite surface change can help prioritise investigation. It is not site-level methane attribution or proof of a leak.

Current status: product and workflow in development.

Is there a field decision buried in your data?

Discuss your project

TenetProduct explorationStreet scale

Places get more useful when we keep their context.

Tenet explores private city intelligence: places saved with the reason they mattered, and findable later by asking a plain question.

Designed exploration, fictional places

Illustrative three-step diagram over a fictional street plan: a place is captured as a node, an intention links it to nearby nodes, and a later question traverses those links to surface the right place.

Read the case notes

What this exploration tests

The concept asks how a private knowledge graph and a bounded assistant could make personal place memory useful, without turning location history into a public feed.

These flows express a product direction from an exploratory MVP specification. They should not be read as verified running features.

Current status: designed product exploration in development.

Are you shaping an experience around place and intent?

Discuss your project

From a useful question to a working product.

We define the product, design the experience, and engineer the system behind it. Every stage ends in something you can test.

  1. Define the product

    Name the user, the decision, the evidence and the acceptance test before anyone commits to a build.

    Direction, prototype, build plan

  2. Build the first working system

    Design and ship one useful workflow, the integrations it needs, and the evaluation that keeps it honest.

    Product, system, handover

  3. Develop it further

    Use real feedback to improve usefulness, reliability, and the product decisions behind both.

    Iterations, evaluation, decisions

Questions we are usually asked first.

What does Adaptivecity do?
We are an independent AI design and engineering practice. We build AI products for the physical world, from sensing and geospatial analysis through to the interface where a person makes the decision.
What does AI for the physical world mean?
Software that reads real conditions from sensors, satellites and places, then turns them into a judgement someone can act on. Our scope is sensing, spatial context, machine learning and human-facing software. It does not cover robot control or autonomous machines.
Who do you work with?
Founders and product teams building environmental intelligence, geospatial products, connected hardware, and software for field operations. We also work with teams building location-aware consumer products.
How does an engagement start?
Usually with a bounded, paid discovery. We map the user, the decision, the available evidence and a measurable acceptance test, then plan the build against it. A well-scoped build can start directly.
Where are you based?
Barcelona, Spain, at Tech Barcelona, Pier 1, Plaça de Pau Vila. We work remotely with teams elsewhere.
How do I get in touch?
Email andrea@adaptivecity.com with what you are building, who it serves, and what has to work.

What should your product understand about the world?

Tell us what you are building, who it serves, and what has to work. We will help you shape the next step.

andrea@adaptivecity.com