Artificial Intelligence & Data Fusion

Haedes; Belgique; Hybrid; EUR 70,000 - 100,000; Full time

Job summary

HAEDES in Belgium is seeking an AI/ML & Data Fusion Engineer to turn complex, heterogeneous data into dependable forecasts, early warnings and actionable insights for rivers, coasts, estuaries, oceans and water-related assets.

You will combine Earth Observation, in-situ IoT measurements and numerical-model outputs to develop predictive models, data-fusion methods and analytical components for scalable environmental applications.

Qualifications

• Master's degree or PhD in data science, CS, math, engineering, physics, environmental science or related field.
• 3+ years of relevant experience in data science, ML, predictive modelling or scientific computing.
• Strong programming in Python and C++, with experience in scientific computing and ML ecosystems.
• Experience with supervised and unsupervised learning, statistical modelling and feature engineering.
• Hands-on time-series analysis, forecasting, anomaly detection and predictive modelling.
• Experience integrating heterogeneous datasets with different spatial/temporal resolutions.
• Ability to work with large datasets and choose appropriate storage/processing formats.
• Solid understanding of uncertainty, bias, data leakage, generalisation and performance metrics.
• Ability to explain methods/results to specialists and non-specialists.
• Professional working proficiency in English (Portuguese not required).

Responsibilities

• Develop, train and evaluate supervised and unsupervised ML models for environmental and engineering apps.
• Create data-fusion methods that reconcile satellite, sensor, geospatial and model data across scales.
• Build time-series forecasting, anomaly-detection and predictive-maintenance indicators.
• Design features and hybrid approaches combining domain knowledge, physical models and data-driven methods.
• Develop reproducible data pipelines for ingestion, QC, transformation, training and inference.
• Quantify predictive uncertainty and monitor model performance for trusted outputs.
• Translate models into demonstrations, visualisations and decision-support components for users.
• Contribute to proposals, client projects and partnerships while exploring new ideas.

Skills

Python

C++

Time-series analysis

Machine learning

Data fusion

Uncertainty quantification

English proficiency

Education

Master's degree or PhD in data science, CS, math, engineering, physics, environmental science or related field

Job description

Belgium | 5+ years of experience | Hybrid & international

Fuse observations, sensors and physical models into intelligence that helps water systems anticipate change.

The opportunity

HAEDES is looking for an AI/ML & Data Fusion Engineer to strengthen our team in Belgium. We seek someone who can turn complex, heterogeneous data into dependable forecasts, early warnings and actionable insights for rivers, coasts, estuaries, oceans and water-related assets.

You will combine Earth Observation, in-situ IoT measurements and numerical-model outputs to develop predictive models, data-fusion methods and analytical components for scalable environmental applications. Working between applied research, engineering and software development, you will help advance the data-driven intelligence within HAEDES’ digital twins and regenerative blue-economy projects.

What you will shape?

• Develop, train and evaluate supervised and unsupervised machine-learning models for environmental and engineering applications.
• Create data-fusion methods that reconcile satellite, sensor, geospatial and numerical-model data across different scales, frequencies and uncertainties.
• Build time-series forecasting, anomaly-detection and predictive-maintenance or risk-indicator solutions.
• Design features and hybrid approaches that combine domain knowledge, physical models and data-driven methods.
• Develop reproducible data pipelines for ingestion, quality control, transformation, training, validation and operational inference.
• Quantify predictive uncertainty, monitor model performance and communicate when outputs are - and are not - fit for use.
• Translate models into clear demonstrations, visualisations and decision-support components for technical and non-technical users.
• Contribute to R&D proposals, client projects and partnerships while exploring promising ideas of your own.

What you bring?

• A master's degree or PhD in data science, computer science, mathematics, engineering, physics, environmental science or a closely related field.
• At least 3 years of relevant experience in data science, machine learning, predictive modelling and/or scientific computing.
• Strong programming skills in Python and C++, with practical experience in scientific-computing, data-science and machine-learning ecosystems.
• Experience with supervised and unsupervised learning, statistical modelling, feature engineering and robust model validation.
• Hands-on experience with time-series analysis, forecasting, anomaly detection and predictive modelling.
• Experience integrating heterogeneous datasets with different spatial and temporal resolutions.
• The ability to work efficiently with large datasets and select appropriate storage, processing and exchange formats.
• A sound understanding of uncertainty, bias, data leakage, generalisation, performance metrics and model limitations.
• The ability to explain analytical methods and results to both specialist and non-specialist audiences.
• Professional working proficiency in English. [Confirm whether Portuguese is required or preferred.]

Additional experience we value

• Environmental, geospatial, satellite, oceanographic, hydrological or IoT sensor data.
• Numerical or physical modelling and hybrid, physics-informed or surrogate-modelling approaches.
• Cloud-based and distributed data or ML workflows, including containers, orchestration and scalable inference APIs.
• MLOps practices such as experiment tracking, versioning, automated testing, deployment, monitoring and retraining.
• Deep learning, spatiotemporal modelling, computer vision, Bayesian methods or data assimilation.
• Forecasting and early-warning services, asset-specific risk analytics or digital-twin applications.

How you work?

You are adventurous enough to explore the unknown and rigorous enough to make the result trustworthy. You work autonomously without working alone, challenge assumptions and invite scientists, engineers, software developers and domain experts into the process. You combine creativity with responsibility, treat uncertainty honestly and keep learning as methods, data and environmental questions evolve.

What makes HAEDES different?

HAEDES blends engineering and soul: logic with emotion, knowledge with wisdom, and process with patterns. We address root causes rather than symptoms and co-create solutions inspired by nature. Our values – nature, trust, connection, creativity, autonomy and authenticity – guide how we choose projects, work with partners and make decisions.

• A living-system organisation that adapts through co-creation, learning and distributed responsibility.
• Freedom to pioneer and pursue original ideas, paired with accountability for data quality and useful outcomes.
• Work at the intersection of water, AI, Earth Observation, numerical modelling, climate resilience and regenerative design.
• A multidisciplinary international network and a nurturing culture with serious expertise, personal guidance and room for humour.

What we offer

• A varied role with meaningful R&D and client work, plus genuine influence on our predictive-analytics and digital-twin capabilities.
• A flexible working rhythm: remote, office-based and on project locations, with occasional travel to Belgium and international clients.
• Remuneration aligned with your experience, skills and responsibilities.
• A personal development plan covering technical growth and self-development.

We look forward to discovering not only what your models can predict, but what they can help us change.