Machine Learning Engineer
New
A
AndersenAdtech, e-commerce
Locations: Warszawa, Rondo Daszyńskiego, 1, Germany, remote, Hungary, remote, Lithuania, remote, Romania, remote, Warszawa, Country code: PLFull-TimeMiddle
Salary18,000 - 20,000 PLN per month
Apply NowOpens the employer's application page
Job Details
- Languages
- Level of English – from Upper-Intermediate and above.
- Experience
- 5+ years.
- Required Skills
- PythonSQLApache AirflowGCPBigQuery
Requirements
- Have 5+ years of hands-on machine learning engineering experience using Python, including developing, deploying, and maintaining production-grade ML models.
- Have experience with XGBoost, CatBoost, or LightGBM, including hyperparameter tuning, feature importance analysis, and advanced model evaluation.
- Have strong SQL and BigQuery skills, including complex feature engineering queries, scheduled data transformations, and data validation.
- Have experience deploying and maintaining ML workloads on GCP, including Cloud Run, Airflow or Composer, Vertex AI, and Pub/Sub.
- Understand model calibration, business-oriented loss functions, and the effects of prediction errors on downstream optimization systems.
- Have practical experience integrating ML outputs with external systems through APIs.
- Have Upper-Intermediate English or above.
- Experience in e-commerce or DTC, particularly LTV prediction, purchase forecasting, churn modeling, or customer segmentation, is preferred.
- Familiarity with BG/NBD, Gamma-Gamma, similar probabilistic customer models, or Bayesian statistics is preferred.
- Experience with advertising platform integrations such as Google Ads API or Meta Conversions API is preferred.
- Knowledge of identity resolution and entity matching techniques is preferred.
- Experience with Terraform, CI/CD pipeline design, Docker, causal inference, A/B testing, or multi-armed bandit optimization is a plus.
Responsibilities
- Build and maintain pLTV models using XGBoost, CatBoost, and BG/NBD within a tiered scoring architecture.
- Design and maintain model calibration layers using normalization methods, haircut scalars, and asymmetric loss functions.
- Develop Gold-layer feature tables in BigQuery from Silver-layer data, including RFM metrics, behavioral embeddings, category affinity, and purchase velocity.
- Evaluate feature completeness and missing-value rates, excluding features with high missing rates among new users.
- Monitor model drift and distribution shifts, log residuals, compare cohorts, and identify causal contamination in retraining data.
- Maintain BigQuery Union-Find identity resolution algorithms connecting GA4 behavioral identifiers with Shopify transactional identifiers.
- Develop production-grade Python applications on Google Cloud Run Jobs and orchestrate workflows through Apache Airflow.
- Deliver ML-generated signals to advertising platforms and maintain automated testing, CI/CD pipelines, idempotent processing, and dry-run safety mechanisms.
View Full Description & ApplyYou'll be redirected to the employer's site
