MLOps Engineer (SDE 3)

Role Overview

We are seeking an experienced MLOps Engineer to design and enhance infrastructure for serving machine learning models. The role focuses on optimizing inference latency and cost efficiency while ensuring robust deployment and monitoring of ML models in enterprise environments.

Core Responsibilities

• Design and refine infrastructure to serve ML models emphasizing low latency and cost-effective operations.
• Develop inference pipelines that effectively balance speed, throughput, and expenses leveraging various acceleration technologies.
• Create systems for monitoring and observability of machine learning workflows.
• Partner with ML Engineers to define and implement best practices for optimized model deployment.
• Deliver scalable and cost-aware solutions fit for enterprise requirements.
• Engage in cross-functional teamwork with ML Engineers, QA, and DevOps teams to continuously improve systems.
• Assess and integrate new tools and technologies relevant to ML system deployment.
• Participate in architectural decision-making related to distributed machine learning systems.

Required Experience and Skills

• Minimum 5 years of professional experience in software development with Python.
• Hands-on expertise with machine learning frameworks, particularly PyTorch.
• Proficiency in optimizing ML models using hardware acceleration tools such as AWS Neuron, ONNX, and TensorRT.
• Experience working with AWS ML products including SageMaker, Inferentia, and Trainium, and managing hardware-accelerated instances.
• Strong background in building and maintaining AWS serverless architectures.
• In-depth knowledge of event-driven architectures and messaging services like SQS and SNS, plus serverless caching techniques.
• Familiarity with containerization technologies like Docker and orchestration platforms.
• Excellent grasp of RESTful API design and development.
• Ability to write secure, high-quality code and knowledge of static analysis tools.
• Good command of English for clear communication, both spoken and written.
• Solid foundation in computer science principles including algorithms, problem solving, and complexity analysis.

Preferred Qualifications

• Experience with ML model compilation, quantization, and profiling to enhance inference performance.
• Background in working within regulated domains that enforce strict compliance on cloud-native deployments.