SIXT is looking for a Senior Machine Learning Engineer to join our team of machine learning experts. You will develop and implement cutting-edge models specifically tailored for price optimization, working closely with data scientists to deploy low latency services in production with the objective to infer optimal bid price, leveraging theories of opportunity cost.
What You'll Do
- Develop advanced regression prediction models to determine optimal prices under capacity constraints, ensuring architectures capture demand elasticity, capacity limits, and temporal stationarity.
- Prototype and evaluate novel architectures, including deep MLPs, transformer-based models for tabular/sequence data, CNNs for image inputs, and embedding-rich hybrids—optimized for accuracy, robustness, interpretability, and transfer learning to maximize out-of-sample generalization.
- Conduct large-scale experimentation and simulation to assess model generalization and analyze behavior across geographies, seasons, and capacity constraints.
- Design and optimize large-scale model training pipelines, leveraging distributed frameworks (PySpark, Dask, Polars) and multi-GPU/multi-node parallelization for efficient deep learning at scale, ensuring reproducibility, fault tolerance, and cost-efficient scaling on AWS.
- Build high-performance data engineering pipelines for terabyte-scale datasets using Polars, Dask, or Spark, implementing columnar storage, memory mapping, and data sharding to enable efficient GPU utilization.
- Implement monitoring and continuous optimization, automating feedback loops, retraining triggers, and hyperparameter tuning (Ray Tune, Optuna), while documenting infrastructure design and mentoring peers on best practices for distributed ML and production-scale architectures.
What We're Looking For
- Deep Learning Modeling: Demonstrated ability to innovate in model design (embeddings, deep MLPs, hybrid architectures) and proven experience with sequence models such as LSTMs, GRUs, and Temporal Convolutional Networks.
- Scalable ML Engineering: Expertise in building and deploying large-scale ML systems handling 50M+ data points or terabyte-scale datasets, leveraging distributed training and GPU acceleration.
- ML Frameworks & Parallelization: Strong proficiency with PyTorch or TensorFlow for large-scale model development and training; skilled in optimizing performance across multi-GPU or distributed environments for datasets exceeding 50M rows.
- Advanced Data Processing: Hands-on experience with PySpark, Dask, and Polars for distributed data preparation and transformation; knowledge of Arrow-based columnar computation and vectorized data pipelines.
- Cloud & HPC: Proficiency with AWS, Kubernetes, and containerized GPU workloads (NVIDIA Docker); experience orchestrating multi-GPU/multi-node clusters for deep learning.
- MLOps & Lifecycle Management: Familiarity with MLflow, Airflow, or Kubeflow for managing large-scale model lifecycles, experiments, and retraining pipelines; strong foundations in deploying and maintaining ML models in production environments and ensuring collaboration between data science and infrastructure teams.
Technical Stack
- ML Frameworks: PyTorch, TensorFlow
- Distributed Data: PySpark, Dask, Polars
- Cloud & Infrastructure: AWS, Kubernetes, NVIDIA Docker
- MLOps & Orchestration: MLflow, Airflow, Kubeflow, Ray Tune, Optuna
Benefits & Compensation
- 28 days of vacation
- Additional day off for your birthday
- 1 volunteer day per year
- Hybrid working model
- Flexible working hours
- No dress code
- Discounts on SIXT rent, share, ride, and SIXT+
- Partner discounts
- Training programs, external conferences, and internal dev & tech talks
- Private health insurance
- Coverflex advantage system
Work Mode
This role follows a hybrid working model.
SIXT is an equal opportunity employer.
