About the role

Frame and solve diverse ML problems.
Build models across the algorithmic spectrum.
Apply NLP / DL to unstructured data.
Design experiments and A/B tests.
Own the full lifecycle of model development.

Skills the posting asks for

Python
NumPy
pandas
scikit-learn
TensorFlow

Summarised from the employer’s posting, which is reproduced in full below.

The employer’s full posting

🌟 Join Impress – Europe’s Leading Health-Tech Innovator!

We're looking for a strong Machine Learning Engineer with 3+ years of hands-on experience and with deep fundamentals in ML algorithms and modeling. You'll design and ship models that drive decisions across our business — scoring, ranking, uplift, forecasting, recommendation, and NLP — owning each problem from formulation through production and measured impact. Our Data & ML team builds the models that power patient and operations decisions at Impress, Europe's largest orthodontic clinic chain, mining signal from patient communications. You'll have room to take these further and to open up new modeling directions as the business grows.

📢 Why we’re cool:

Work with an international and multicultural team
Competitive salary
Teeth aligner and whitening benefits
Collaborative work environment and positive culture
Opportunities to grow within a fast-paced, innovative company and real start-up experience with big challenges
Fresh fruits and healthy snacks at the office

🔥What You’ll Do:

Frame and solve diverse ML problems — classification, regression, ranking, uplift / causal modeling, forecasting, recommendation, anomaly detection, and some NLP processing.
Build models across the algorithmic spectrum — from gradient boosting and classical ML to deep learning (mostly inference) — choosing the right tool, not the trendy one.
Apply NLP / DL to unstructured data (text, conversations, communications): classification, intent detection, embeddings, summarization, information extraction.
Design experiments and A/B tests — define offline metrics and online success criteria, reason about baselines, causal effects, and statistical significance, and prove that models actually move the needle.
Own the full lifecycle — data extraction and feature engineering, training and evaluation, deployment, retraining, and monitoring for drift and data quality.
Set the technical bar — bring rigor to evaluation, guard against leakage and overfitting, and mentor on solid ML practice.

🔥Requirements:

Strong ML fundamentals: probability and statistics, optimization, bias/variance, regularization, model evaluation, and a real understanding of the algorithms behind the libraries.
Breadth of modeling experience: tree ensembles (boosting/bagging), linear models, clustering, and deep learning (CNNs/RNNs/transformers) — and the judgment to choose between them.
Experimentation: Familiarity with uplift / causal inference and experimentation, or a strong drive to master it.
NLP / LLM experience (embeddings, transformers, fine-tuning or prompting).
Technical Stack: Strong Python and the ML ecosystem (NumPy, pandas, scikit-learn; PyTorch or TensorFlow; gradient-boosting libraries).
Production Track Record: shipping models to production, not just notebooks — and measuring their impact.
Software Fundamentals: clean code, SQL, version control, testing.

💪Nice to have:

MLOps maturity: experiment tracking, CI/CD for models, feature stores, model monitoring.
Infrastructure: Cloud (AWS / GCP), data warehouses, orchestration (Airflow or similar), serving (FastAPI, Docker).
Community: Publications, competitions.

At Impress we cultivate a culture of inclusion and diversity. We celebrate our employees' individual strengths, views, and experiences and we encourage all candidates to apply, without regard to race, color, religion, gender identity, sexual orientation, age, national origin, disability, or any other factor.🌈💪