Senior Machine Learning Engineer - Research Optimisation
Core
Bridge research and production by hardening experimental models, optimizing training/inference performance, and building shared ML libraries and tooling for Canva's design platform.
Role type
Senior Machine Learning Engineer (Research Enablement & Optimization)
Builds
Production-ready ML models, optimized training/inference pipelines, shared libraries, SDKs, and rollout frameworks for multimodal and generative AI features.
Domain
Design technology, Generative AI, Multimodal models, High-performance computing
Deliverable
production ML models | infrastructure | product features
Required skills
Python, PyTorch, Kubernetes, CI/CD, Distributed training (FSDP/DDP/DeepSpeed), GPU profiling, Observability, Cloud platforms (AWS)
Preferred skills
CUDA kernel optimization, Model serving tools (ONNX/TorchScript/Triton), High-performance storage systems (Weka/Vast/Lustre), MLOps practices
Technologies
PyTorch, Kubernetes, AWS, FSDP, DDP, DeepSpeed, ONNX, TorchScript, Triton, Weka, Vast, Lustre
Responsibilities
Refactor and containerize research experiments for production deployment; Profile and optimize PyTorch training jobs for GPU efficiency; Build shared libraries and SDKs to standardize interfaces; Implement CI/CD workflows and artifact management for ML services; Establish observability and reliability practices (metrics, tracing, load testing); Optimize inference latency and cost via batching, quantization, and hardware utilization.
Seniority
Senior, hands-on IC