Senior Engineer AWS AI & MLOps

At Aptiv, the team focuses on new mobility by making transportation safer, greener, and more connected through software capabilities, advanced computing platforms, and networking architectures. The role involves working on AI & MLOps infrastructure for L2++ autonomous driving systems in complex urban environments, ensuring automotive-grade reliability, safety, and performance standards.

Technical Skills:

• Strong experience with AWS cloud architecture for ML workloads
• Hands-on expertise in multi-GPU training (Ray or equivalent), EKS / Kubernetes, Terraform, Airflow, MLflow
• Proficient Python programming for ML and platform automation
• Experience building and operating CI/CD pipelines using GitHub

Machine Learning Competence:

• Deep understanding of ML training pipelines including data ingestion, preprocessing, quality assurance, and train/test validation
• Experience supporting large-scale ML experimentation and productionization

Automotive & Algorithmic Understanding:

• Solid understanding of perception systems and behavioral/decision-making algorithms
• Prior experience in automotive, ADAS, or autonomous driving environments required
• Familiarity with constraints of safety-critical and real-time systems

MLOps & Cloud Architecture:

• Design and own end-to-end MLOps architecture on AWS for autonomous driving workloads
• Architect multi-zone, highly available ML platforms supporting urban L2++ hands-off use cases
• Build and operate scalable multi-GPU training environments using Ray clusters
• Define infrastructure standards for compute management, networking, storage, and security

AWS Platform & Infrastructure:

• Implement and manage Amazon EKS / Kubernetes for ML workloads
• Manage VPC architecture, subnets, routing, and network isolation
• Use S3 Intelligent-Tiering for cost-efficient storage of large-scale sensor and training data
• Use AWS Lambda for event-driven ML workflows and automation
• Provision AWS IoT infrastructure via Terraform
• Ensure multi-zone resilience, fault tolerance, and disaster recovery strategies

MLOps Pipelines & Tooling:

• Design and operate ML pipelines using Apache Airflow and MLflow
• Implement CI/CD pipelines for ML and infrastructure using GitHub
• Enable reproducible, traceable, and auditable ML workflows aligned with automotive standards

Machine Learning Engineering:

• Enable scalable data ingestion and processing pipelines for sensor-rich datasets
• Establish data quality checks, validation frameworks, and train/test split governance
• Support ML teams with optimized workflows for training, evaluation, and deployment
• Collaborate on best practices for training at scale, including performance tuning and cost optimization

Algorithmic & Domain Collaboration:

• Work closely with ML researchers and engineers on perception and behavioral algorithms
• Translate algorithmic requirements into production-ready infrastructure
• Apply automotive domain knowledge to ensure platform suitability for safety-critical systems

Observability, Scalability & Operations:

• Build strong monitoring, logging, and observability for ML systems and infrastructure
• Enable performance metrics, failure detection, and operational insights across the ML lifecycle
• Continuously improve platform scalability, reliability, and operational efficiency