System Software Engineer - AI Infra (Hybrid Montreal)

Delos Data; Montreal; Hybride; CAD 120 000 - 180 000; Plein temps

Avantages offerts par ce poste

Equity

Benefits

RRSP

Résumé du poste

Delos Data in Montreal is seeking a System Software Engineer to redefine the infrastructure layer for large-scale AI. You will bridge high-level AI frameworks and low-level system software, designing communication and execution primitives for running models across thousands of GPUs.

You should be fluent in C++ and Python, with strong Linux and OS internals knowledge, and experience optimizing distributed workloads.

Qualifications

• Proficiency in C++ and Python with systems programming fundamentals.
• Strong Linux development experience and OS internals knowledge.
• Experience optimizing distributed workloads and identifying bottlenecks.

Responsabilités

• Collaborate across the stack to influence foundational technology design.
• Identify and resolve performance bottlenecks in distributed training/inference workloads.
• Conduct rigorous performance benchmarking on multi-node clusters.

Connaissances

C++

Python

Systems programming

Memory management

Concurrency

OS internals

Linux development

Formation

Bachelor's or Master's degree in Computer Engineering, Computer Science, or related field

Outils

CUDA

PyTorch

JAX

DeepSpeed

vLLM

Dynamo/TRT-LLM

Telemetry tools

Description du poste

Delos Data in Montreal is seeking a System Software Engineer to redefine the infrastructure layer for large-scale AI. You will bridge high-level AI frameworks and low-level system software, designing communication and execution primitives for running models across thousands of GPUs.

You should be fluent in C++ and Python, with strong Linux and OS internals knowledge, and experience optimizing distributed workloads.