SDE III - MLOps

About the role

Join Jumio, a leader in identity verification, to build the foundation for scalable and efficient machine learning model serving. You will play a key role in enhancing our ML infrastructure and developer experience.

Key facts

What you'll do

• Improve systems for managing ML assets like models and data, focusing on developer usability and governance.
• Design and optimize infrastructure for deploying and serving ML models, prioritizing inference speed and cost efficiency.
• Create efficient inference pipelines that balance speed, volume, and cost across different acceleration technologies.
• Develop enterprise-scale solutions that are cost-effective.
• Work collaboratively within a distributed, cross-functional team to continuously improve systems.
• Partner with Machine Learning Engineers, Quality Assurance Engineers, and DevOps Engineers.
• Research and integrate new technologies and tools.
• Contribute to architectural decisions for distributed ML systems.

Requirements

• A minimum of 5 years of software engineering experience, with a strong focus on Python.
• Experience managing the machine learning model lifecycle using tools like MLFlow or Weights & Biases.
• Familiarity with data management concepts, including quality, transformation, and cataloging.
• Experience with ML frameworks, especially PyTorch.
• Experience optimizing ML models for hardware acceleration using technologies such as AWS Neuron, ONNX, or TensorRT.
• Proven background in building and operating AWS serverless architectures.
• Solid understanding of event-driven processing, SQS/SNS, and serverless caching.
• Experience with containerization using Docker and orchestration tools.
• Strong skills in designing and implementing RESTful APIs.
• Proficiency in writing high-quality, secure code and using static code analysis tools.
• Excellent analytical, conceptual, and communication abilities in English.
• Demonstrated ability to apply computer science fundamentals in algorithm design, problem-solving, and complexity analysis.

Nice to have

• Experience with model compilation, quantization, or performance profiling and benchmarking of ML inference systems.
• Experience working in regulated industries with strict compliance requirements for cloud solutions.

Skills & tools

• Python
• MLFlow / Weights & Biases
• PyTorch
• AWS Neuron / ONNX / TensorRT
• AWS Serverless (SQS, SNS)
• Docker
• Kubernetes
• RESTful APIs

Practical notes

• Jumio is committed to equal opportunities and welcomes applications from all backgrounds.
• Jumio's Applicant Privacy Notice outlines how your data will be used during the application and hiring process.