Machine Learning Engineer

Project: Advanced Simulation & Decision-Support for Hydrology
Location: Remote
Start: ASAP

About the role

We’re hiring a Senior ML Engineer to build forecasting and decision-support models for floods, droughts, and compound extreme events. You’ll design neural models (LSTM/RNN/CNN), develop prescriptive simulations (mass balance & hydrological processes), and deliver transparent, benchmarked results stakeholders can trust.

Responsibilities

• Predictive modeling: Design, train, and tune NN models (LSTM/RNN/CNN) for hydrological time series and basin/climate data.
• Prescriptive modeling & simulation: Implement mass-balance and process-based models for system behavior and what-if scenarios.
• Scenario analysis: Forecast floods/droughts/extremes; benchmark against historical & observed datasets.
• Optimization & decision support: Use Pyomo, Gurobi, SimPy and metaheuristics for robust decisions under uncertainty.
• Geospatial integration: Ingest GIS, raster/vector, and climate datasets for real-time simulations.
• Explainability & uncertainty: Add XAI and uncertainty quantification layers to outputs.
• Transparency & docs: Maintain clear documentation, benchmarks, and reproducible pipelines.

Requirements

• Solid track record with Neural Networks, LSTM, and time-series modeling.
• Hands-on with Pyomo / Gurobi / SimPy (or metaheuristic optimization).
• Knowledge of mass-balance modeling, hydrological processes, and climate data systems.
• Python + ML stack: TensorFlow / PyTorch / scikit-learn.
• Geospatial skills (GIS; raster/vector data).
• Experience with explainability (XAI) and uncertainty assessment.
• Strong analytical writing and documentation.

Nice to have

• Background in hydrology / water resources / climate modeling.
• Experience with prescriptive analytics & decision-support for environmental domains.

Tech stack (core)

Python, PyTorch/TensorFlow, scikit-learn, Pyomo, Gurobi, SimPy, NumPy/Pandas, xarray, rasterio, GDAL, GeoPandas, MLflow/DVC, Docker.

What we offer

• Impactful work in climate & water-risk mitigation.
• Freedom to shape modeling approaches and benchmarks.
• Support for publications, conferences, and open science where possible.
• Flexible hybrid setup.