Member of Technical Staff, ML Systems

Location: Remote from São Paulo, Brazil

Company

hirify.global builds machine learning systems and an internal platform for developing, governing, and deploying models across cloud and customer environments.

What you will do

• Build CUDA kernels and compute primitives for training and serving graph neural networks.
• Evolve the Monad sampler and distributed-training library, including neighbor sampling and training performance improvements.
• Define binary data formats and own materializations and feature backfills for training and evaluation.
• Establish data contracts and consumption requirements with teams producing customer and proprietary datasets.
• Build experiment tracking, checkpointing, evaluation, model registry, lineage, and versioning infrastructure.
• Define release gates and ensure production, batch, and on-premise models have governed, reproducible release records.

Requirements

• Strong systems engineering skills and production-quality Python.
• Experience with distributed training, such as Ray or PyTorch Distributed, and multi-node GPU workloads.
• Experience with columnar data formats and large-scale data materialization.
• Familiarity with experiment tracking, model registries, evaluation, and reproducibility tooling.
• Product mindset and ability to build internal platforms for researchers and platform engineers; data science experience is not required.

Nice to have

• CUDA kernel development or GPU performance optimization.
• Graph neural networks or graph sampling at scale.
• Lance, Arrow, or other columnar and indexed storage formats.
• Multi-cloud GPU compute, such as SkyPilot.
• Model governance or audit experience in financial services or other regulated environments.