You will work as part of an established and growing research team, and help to develop frameworks that will be used to train large models on a distributed cluster. You’ll closely collaborate with researchers, platform and software engineers, and leverage your knowledge of infrastructure and training libraries to accelerate the rollout of a broad array of models.

Your Core Responsibilities:

• Develop & support libraries for:
  • Accelerating training on a distributed environment
  • Automating ML workflows, experiment evaluation and management, hyperparam tuning (autoML) & retraining
  • The efficient use of GPU hardware in research cluster (data pipeline design, leveraging libraries such as GPUDirect & DALI)
• Evaluate and roll out third-party tooling (e.g. MLflow; Neptune; Ray; …)
• Dig into the internals of open-source ML tools to extend their capabilities and fix fundamental bugs
• Leverage experience in scheduling tools (k8s, slurm etc) to support the effective use of contested resource across a broad research team

Your Skills and Experience:

• MS degree in CS or similar fields or equivalent experience
• 3+ years of relevant work experience
• Experience in fundamental ML frameworks like Pytorch and Tensorflow
• Experienced in Python, experience in C++ is highly desired