
Member of Technical Staff β ML Systems
Posted 3 days ago

Posted 3 days ago
This is a fully remote position, open to applicants in Brazil.
β’ Develop CUDA kernels and computational primitives for the training and deployment of graph neural networks (GNNs)
β’ Enhance Monad, the library for sampling and distributed training, focusing on neighbor sampling and training efficiency
β’ Define binary data formats (Lance, Arrow, CSR/CSC) and manage materializations and feature backfills for both training and evaluation
β’ Establish data contracts and consumption criteria with teams that work on customer and proprietary datasets
β’ Create and maintain infrastructure for experiment tracking, checkpoints, and evaluations, ensuring reproducibility by design
β’ Oversee the model registry, including lineage tracking, versioning, and compatibility across models, embeddings, and downstream applications
β’ Set and execute release criteria for managed production, batch, and on-premise model deployments
β’ Facilitate auditing of data, code, configuration, and proof associated with each release
β’ Enhance time-to-experiment, time-to-governed-release, GPU training throughput, release documentation, lineage, and reproducibility
β’ Proficient in systems engineering and production-level Python programming
β’ Experience in distributed training frameworks (e.g., Ray, PyTorch distributed) and multi-node GPU environments
β’ Knowledge of columnar data formats and large-scale data materialization techniques
β’ Familiarity with machine learning lifecycle tools: experiment tracking, model registries, evaluation, and reproducibility
β’ A product-oriented approach: view an internal platform as a product serving real users
β’ Experience in CUDA kernel development or optimizing GPU performance (preferred)
β’ Understanding of graph neural networks or large-scale graph sampling (preferred)
β’ Knowledge of Lance, Arrow, or similar columnar/indexed storage formats (preferred)
β’ Experience with multi-cloud GPU computing solutions (e.g., SkyPilot) (preferred)
β’ Understanding of model governance or audit requirements, particularly in financial services or other regulated sectors (preferred)
β’ Competitive salary and performance-based bonuses
β’ Comprehensive health, dental, and vision insurance plans
β’ Flexible working hours and remote work options
β’ Opportunities for professional development and career advancement
β’ Supportive and inclusive company culture
Rich Products Australia
Coursedog
DaCodes.
DaCodes.
Get handpicked remote jobs straight to your inbox weekly.