Data, forecasting and geospatial

I turn sensors, weather forecasts and satellite imagery into decisions: predictions on sensor-measured quantities, recommendation engines, operational maps from satellite imagery, scoring engines. I choose the model on the numbers, often a well-built Random Forest beats a transformer, and deliver it as a service with APIs, database and monitoring.

Typical problems

  • IoT and weather sensors that produce data but no usable forecast.
  • Satellite imagery to turn into operational maps for people working on the ground.
  • Domain or business rules to encode into a repeatable, verifiable engine.
  • Deep learning models proposed out of habit, without comparison to a baseline.
  • Large geospatial data (land registry, maps) that must be served reliably.

What I deliver

  • Feature engineering on time series (lags, windows, physical balances) with group-wise validation.
  • Classical and deep models (Random Forest, Temporal Fusion Transformer) compared with documented metrics.
  • Sentinel-2 pipelines: spectral indices, cloud masking, zoning, GeoJSON and Shapefile export.
  • FastAPI services with InfluxDB, MariaDB or PostGIS, Keycloak authentication, Helm and CI/CD.
  • ONNX export and scheduling of forecasting jobs.
  • Technical documentation and reports for decision-makers.

How I approach it

First a baseline and a shared metric, then the model. Every prediction carries its expected error. Jobs run on a schedule, write to the database and are monitored from a dashboard.

Email meLinkedIn

Work in this area

Precision agriculture · 2025-2026

Irrigation recommendation and soil forecasting

Problem
Field sensors, weather forecasts and crop phenology to turn into a daily decision: how much to irrigate and fertilize, and when.
What I did
An agronomic engine with soil hydrology, N-P-K chemistry, BBCH phenological stages and evapotranspiration, on MariaDB and InfluxDB with Keycloak authentication. Multi-target soil state forecasting with Random Forest and over one hundred features (average SMAPE 4.7% across five targets). A Temporal Fusion Transformer evaluated with Optuna and ONNX export, then dropped: the classical model performed better.
  • scikit-learn
  • Random Forest
  • TFT
  • ONNX
  • InfluxDB
  • FastAPI
  • Keycloak

Precision agriculture · 2026

Prescription zones from Sentinel-2 imagery

Problem
An agronomist draws a field boundary and wants vigor maps and variable-rate zones without GIS or satellite expertise.
What I did
A service that downloads Sentinel-2 imagery from Copernicus for a polygon and date range, computes NDVI with cloud masking, groups it into vigor zones with K-Means and exports GeoJSON, Shapefile and ISO-XML for machinery, plus georeferenced overlays. A React frontend with Leaflet to draw boundaries and pick scenes, a mock mode with no credentials, 141 tests.
  • rasterio
  • Sentinel-2
  • NDVI
  • K-Means
  • React / Leaflet
  • FastAPI

Frequently asked questions

What kind of projects do you work on?

Industrial computer vision projects, LLM-based systems (RAG, agents, structured extraction), forecasting on IoT and geospatial data, and web and mobile platforms that turn these models into a product. From PoC to production, with code, tests and deployment.

How long does a prototype take?

A measurable PoC generally takes two to six weeks, depending on the available data. Production depends on the integration: one to three months for a service with APIs, tests and deployment.

How does pricing work?

A tailored quote after the call, project-based or by the day, or a monthly retainer for those who need an ongoing technical reference. No hidden costs: tools, licenses and infrastructure are decided together and remain yours.