Staff Machine Learning Engineer (Platform)
Core
Building and operating the infrastructure and systems that enable Neara's ML discipline to move fast, ship reliably, and scale without breaking for engineering-grade digital twins of electricity grids.
Role type
Staff Machine Learning Platform Engineer
Builds
Training pipelines, serving architecture, experiment management, and monitoring systems for multi-modal spatial frontier models
Domain
Energy infrastructure + Geospatial AI + Distributed Systems
Deliverable
infrastructure
Required skills
Distributed systems design, CUDA kernel optimization, Production ML platform architecture, System design, Python, PyTorch, Cloud platform expertise, Kubernetes/Docker, Team leadership
Preferred skills
Experience with sparse tensor implementations, Geospatial data handling, On-prem or neocloud environments
Technologies
Python, PyTorch, CUDA, AWS, GCP, Azure, Kubernetes, Docker
Responsibilities
Define multi-year technical roadmap for training pipelines and serving architecture, Develop foundational tooling to accelerate ML delivery from idea to production, Solve hard distributed systems problems for training across varied GPU hardware, Design scalable serving architecture to handle spiky production loads, Unblock the ML team by defining contracts between training, evaluation, and serving, Mentor and coach ML engineers to raise the technical bar
