Staff Machine Learning Engineer (Platform) - Australia based - Full Relocation Provided at Neara

on-site · full_time · Visa sponsorship

Apply for this role at Neara

Responsibilities:
- Define and own the end-to-end ML platform roadmap: training pipelines, model serving, experiment management and monitoring.
- Build developer-facing tooling to shorten cycle time from prototype to production and standardise workflows.
- Design and operate scalable distributed training and serving systems that respect data residency and security constraints.
- Optimise performance across varied GPU hardware (including CUDA and sparse tensor implementations) and orchestrate multi-node training/serving on Kubernetes.
- Ensure reliable deployment and monitoring for frontier spatial models used by global utility customers; collaborate with research, product and infra teams.
- Mentor engineers on MLOps best practices and elevate platform reliability and cost-efficiency.

Requirements:
- 10+ years engineering experience with demonstrable ownership of platform or infrastructure systems.
- Strong Python and PyTorch experience; solid exposure to CUDA and performance optimisation for training/serving.
- Proven experience with Kubernetes, Docker, cloud providers (AWS/GCP/Azure) and containerised deployments.
- Deep understanding of distributed systems, system design, model serving, training pipelines, experiment tracking and observability.
- Experience with geospatial/point-cloud workflows, data unification or similar uncommon data types is highly desirable.
- Excellent cross-team communication and experience delivering production ML to customers; willingness to relocate to Sydney with provided sponsorship.