Senior Distributed Machine Learning Engineer

Overview

We are looking for an experienced senior distributed machine learning (ML) engineer to join our team dedicated to pioneering AI safety research. This role involves close collaboration with ML scientists to tackle complex challenges in training and inference using large-scale models.

Key Responsibilities

• Partner with researchers to speed up research workflows, model training, and inference, enabling efficient use of distributed computing for large models.
• Identify and resolve performance issues by profiling code from research experiments, debugging problems, and optimizing computational resource usage.
• Create and maintain tools and libraries that streamline and coordinate the utilization of distributed systems for research activities.
• Document and uphold best practices in the development of large-scale distributed ML model pipelines.

Qualifications and Skills

• Bachelor's degree in computer science, computer engineering, or software engineering is mandatory. A master's or PhD focusing on machine learning or distributed ML systems is preferred but not mandatory if the candidate exhibits exceptional skills and experience.
• Minimum three years of hands-on experience designing and implementing frameworks for distributed ML training, with recent expertise in tools such as Megatron, DeepSpeed, HuggingFace Accelerate, Fully Sharded Data Parallel (FSDP), vLLM, and verl.
• Strong teamwork capabilities to collaborate with cross-disciplinary teams, maintain documentation of best practices, and remain current with emerging trends in ML and software development.
• Proficiency using cloud services like AWS, Google Cloud Platform, or Azure, along with workload management frameworks such as Ray or SLURM.
• Familiarity with GPU profiling solutions including PyTorch profiler, PyProf, and NVIDIA Nsight.
• Experience with containerization and orchestration technologies such as gRPC, Docker, and Kubernetes.
• Knowledge of data infrastructure technologies, including vector databases.
• Proven contributions to high-quality research projects in deep learning.

Additional Information

The job title "Engineer" is used here for reference and may differ officially depending on jurisdiction.

Compensation and Benefits

• Chance to impact a significant mission dedicated to safe AI development.
• Extensive health benefits, including mental health and wellness accounts.
• Starting with 20 days of paid vacation per year.
• Employer retirement contributions of 4% without a requirement for employee matching.
• Extra compensation equal to 8% of salary available for retirement savings or bonuses, which is independent of individual or group performance.
• Work alongside a dedicated team of world-class experts.
• An inclusive workplace located in a vibrant office in Little Italy's Mile-Ex district, easily accessible by public transport.

About LawZero

LawZero is a Montreal-based non-profit organization devoted to advancing research and creating technical solutions to promote safe-by-design AI systems, inspired by leading researcher Professor Yoshua Bengio. Our vision is to develop AI that understands the world truthfully through transparent reasoning, enhancing scientific discovery, overseeing agentic AI, and addressing AI risk. We strive to nurture AI as a global public good focused on human flourishing.

Inclusion Statement

We value diversity and foster a working environment marked by fairness, openness, and respect for differences. We encourage applications from highly qualified candidates committed to our mission in an inclusive and collaborative culture.

Privacy Notice

Your personal data will be processed for employment evaluation under our Privacy Policy. You may have rights to access, correct, or erase your personal information according to applicable laws. Details for exercising these rights are provided in our Privacy Policy.