DevOps and AWS Infrastructure Engineer - GPU

Clevanoo LLC is recruiting a contract DevOps and AWS Infrastructure Engineer for a client focused on building a spatial intelligence platform for brand activation and live events. The engineer will manage the cloud execution environment, ensuring efficient processing of multi-camera event footage and delivering analytics to clients.

Responsibilities:

• Own the cloud execution environment end to end
• Design S3 bucket architecture, lifecycle rules, and event notifications that trigger pipeline execution automatically on footage upload
• Set up AWS Batch compute environment targeting GPU instances with appropriate vCPU and memory allocations
• Handle pipeline orchestration, success and failure routing, and compute instance termination using Step Functions state machine
• Manage ECR repository for versioned Docker image management
• Implement least-privilege IAM roles for Batch and Step Functions
• Containerize full Python pipeline using NVIDIA base image with validated GPU passthrough and local execution support
• Route output of processed files to correct S3 prefixes on job completion
• Enforce LOG_LEVEL ensuring no raw spatial data reaches CloudWatch
• Conduct infrastructure dry-run validation and acceptance testing against sample footage

Requirements:

• Production experience with AWS Batch, S3, Step Functions, and ECR
• Strong Docker containerization skills including GPU passthrough and dependency pinning
• Solid IAM and least-privilege role architecture experience
• Familiarity with S3 event-driven pipeline triggers and compute lifecycle management
• Ability to work from a defined scope of work with locked deliverables and explicit acceptance criteria
• 3+ years of work experience with AWS Step Functions
• 5+ years of work experience with Amazon Web Services (AWS)
• Working in a remote setting
• 3+ years of work experience with Docker
• Experience coordinating infrastructure handoff with a CV or ML engineering team