Machine Learning Platform Engineer

Role overview

Build and operate the infrastructure that takes machine learning capabilities from experimentation into dependable production systems. The role spans model development workflows, serving, evaluation, and operations, with a focus on helping engineering and research teams ship models efficiently while meeting performance and cost needs.

Responsibilities

- Design and operate systems for model training, evaluation, deployment, inference, and experimentation. - Develop model serving infrastructure for workloads that require high throughput and low latency. - Build reproducible pipelines for data preparation, training, evaluation, model release, and ongoing improvement. - Create tools that help ML practitioners run experiments, assess model quality, and deliver changes safely. - Establish observability, monitoring, tracing, and alerting for production machine learning workloads. - Identify bottlenecks and improve reliability, scalability, latency, throughput, and infrastructure cost.

Requirements

- Strong software engineering fundamentals and experience building production systems. - Experience with ML infrastructure, platform engineering, or production machine learning systems. - Practical experience in model deployment, inference, evaluation, or data pipelines. - Understanding of distributed systems and reliability concerns. - Ability to write maintainable production code and work through evolving requirements. - Ownership, experimentation, and continuous improvement in a fast-moving environment.

Nice to have

- Familiarity with Python, PyTorch or JAX, and model serving tools such as vLLM, SGLang, or TensorRT-LLM. - Experience with cloud infrastructure, workflow orchestration, GPU performance, vector databases, or retrieval systems.