Einstein Hospital Israelita Albert Einstein is looking for a Data Engineer to implement, optimize, and operate Artificial Intelligence workloads in a hybrid High-Performance Computing environment, combining on-premises and cloud resources. You will ensure massive models can be trained and executed in a distributed, efficient, and secure manner, integrating computational infrastructure, data pipelines, and model development to enable large-scale institutional use cases and research projects.

What You'll Do

Implement and execute multi-GPU and multi-node distributed training.
Optimize the performance of AI workloads regarding memory, communication, throughput, and scalability.
Configure and operate Machine Learning pipelines on Kubernetes/Kubeflow.
Develop processing and loading pipelines for large volumes of data.
Integrate on-premises and cloud environments for hybrid execution.
Support researchers in running models and experiments in the HPC environment.
Automate routines for execution, monitoring, and reproducibility of experiments.
Contribute to operational improvements and efficient use of the cluster.

What We're Looking For

Complete university degree.
Proficiency in Python for Machine Learning and distributed processing applications.
Experience with deep learning frameworks (PyTorch, TensorFlow or similar).
Experience with containers (Docker) and orchestration with Kubernetes.
Fundamentals of distributed computing and parallelism.
Administration of Linux environments.
Handling of large datasets and building data pipelines.
Code versioning and workflow automation.

Nice to Have

Experience with GPU-based HPC environments.
Familiarity with Kubeflow or other MLOps platforms.
Experience with training or fine-tuning Large Language Models and multimodal models.
Knowledge of training optimizations (mixed precision, sharding, checkpointing).
Experience with hybrid infrastructure (on-premises + cloud).
Understanding of high-performance networks and I/O optimization.

Technical Stack

Python, PyTorch, TensorFlow, Docker, Kubernetes, Kubeflow, Linux

Benefits & Compensation

Health care: Cuidar Program, Einstein Clinics, Telemedicine, Pharmacy Plan, Medical Plan, Dental Assistance and In-Company Dental Office.
Well-being: Wellhub (Gympass), TotalPass, Coral, Personal Guidance Program and SESC.
For you and your family: Extended paternity leave, Life Insurance, Daycare or Daycare Assistance for mothers or fathers with legal custody, Assistance for Children with Disabilities and Private Pension with zero fees.
Food: Food Voucher, Meal Voucher or On-site Cafeteria.
Mobility: Transportation Voucher, Shuttle Service, Parking, Ride-sharing App and Metro Shuttle.
Benefits Club: For savings and advantages when purchasing products and services in various categories.
Mais Conectados Program: Remote work in Telework or Hybrid modalities according to activity and area of operation.