Director, Machine Learning Operations

Job Description

Interested in pushing a Kubernetes (AWS EKS) cluster past 10,000 nodes as you run a complex DAG of deep learning based algorithms, to produce petabytes of rich and meaningful data about the physical world? Ensuring rock solid infrastructure and up time as you support imagery and analytics workflows on the biggest natural disasters occurring in the USA, Australia, New Zealand and Canada as they unfold?

This role is the senior people and technical leader for the “Processing Engine” team, which acts as the layer between DevOps and the AI and Computer Vision group at Nearmap. We need a skilled, ops-minded backend software engineer with a passion for applying their skills to enable machine learning and AI., and maturity as a people leader.

Day to day the Processing Engine team develops, maintains and supports large scale backend pipeline infrastructure that must burst to thousands of nodes with impeccable observability and reliability. They enable the workflows for training, deploying and monitoring real time executed ML models on our Ray based realtime cluster, and are building out the platform and tooling for our first LLM based products. Key relationships are to act as the bridge between an AI focussed DevOps team, and the ML Engineers, Data Scientists and other AI specialists in AICV. They’re not the ones to train the models, craft the algorithms and analyse the data directly - but their passion is in building the systems necessary to support this work and its ongoing deployment and success.

The Nearmap AI tech stack has been live in production for over 5 years, crunched tens of petabytes of imagery, and is depended on by a wide range of businesses and organisations to meet critical needs such as disaster response, property risk monitoring and environmental management at continent scale. This is a high impact role where uptime and quality matter to real people and situations.

Key Responsibilities

• Large scale (tens of petabytes and thousands of nodes) batch processing of AI and deep learning workflows
• Real-time ML model deployment and monitoring
• Platform and Tooling to weave LLMs into core products
• Tooling and workflows to support data scientists
• Global Team leadership of ML Ops and Engineering specialists - based in Sydney and Poland, supporting a team and market in Australia and the USA
• Providing the tooling, systems and infrastructure that sits on DevOps’ management of AWS cloud and Kubernetes (EKS, Ray, Terraform, Kubeflow, Karpenter, ArgoCD, WandB, Confluent, etc), and enables data scientists and ML Engineers to do their best work