Senior ML Engineer

Risk Modeling

We are looking for a Senior Machine Learning Engineer for a freelance assignment with a global technology company operating in the Earth observation, geospatial analytics, and risk intelligence domain. The role focuses on building robust, production-grade machine learning systems that turn large-scale real-world data into reliable, defensible risk models.

If you are a senior-level ML engineer who combines strong modeling expertise with production-grade engineering and a focus on statistical rigor, we would like to hear from you.

Start

ASAP

Duration

7months+

Location

Espoo/Hybrid

Allocation

100%

Description

In this role, you will take ownership of the training and calibration infrastructure for advanced machine learning models. You will work closely with data scientists and engineers to scale models from research prototypes into production systems that operate on large datasets and deliver statistically sound outputs.

A key part of the assignment is ensuring that model predictions are not only performant but also well-calibrated and externally defensible, meeting high standards for validation and reproducibility.

You will operate across the full ML lifecycle—from training pipelines and experimentation frameworks to model evaluation, calibration, and deployment readiness—within a modern cloud and HPC-enabled environment.

Start: ASAP / negotiable
Work model: Hybrid: preferred: on-site 3 days per week (Espoo)

Requirements

Master’s degree (or higher) in Computer Science, Machine Learning, Statistics, Applied Mathematics, or a related field
5+ years of hands-on experience training and deploying ML models in production environments
Strong experience with large-scale datasets and distributed or GPU-based training
Proven expertise in probability calibration (e.g. isotonic regression, Platt scaling) and ability to diagnose calibration issues
Deep understanding of evaluation for imbalanced classification, including calibration and ranking metrics
Experience with hyperparameter optimization at scale (e.g. Optuna, Ray Tune)
Solid background in ML/ModelOps practices, including:
- Experiment tracking
- Model versioning
- Reproducibility and artifact management
Strong Python skills (pandas, NumPy, scikit-learn)
Experience with cloud environments (AWS) and distributed workloads
Ability to communicate technical results clearly to non-technical stakeholders
Experience collaborating cross-functionally with Data Engineers and Data Scientists

Nice to have

Experience delivering ML-powered features in production (CI/CD, monitoring)
Understanding of geospatial data challenges (e.g. spatial cross-validation)
Experience with uncertainty quantification methods (e.g. conformal prediction, quantile regression)
Familiarity with HPC environments (e.g. SLURM-based clusters)
Experience with Databricks, PostGIS, or AWS RDS/Aurora
Background in insurance, catastrophe modeling, or climate risk
Familiarity with tabular deep learning approaches (e.g. TabNet, FT-Transformer)

Lue lisää projektin yksityiskohdista kirjautuneena.

Luothan tunnukset vaikka olisit täyttänyt yhteystietolomakkeen vanhoilla sivuillamme, kiitos!