Lead malware detection with transformer models and Python in a fully remote position based in Canada. Build production systems for evolving threats.
This senior role develops, evaluates, and deploys machine learning systems using production malware data. You will work with PyTorch, Hugging Face, SQL, and classical machine learning methods. The work balances detection accuracy, operational costs, and resource constraints. It covers the complete lifecycle from experimentation through monitored production deployment.
Key Responsibilities: Design, train, fine-tune, and deploy malware detection models
Classify malicious files and prioritize suspicious samples
Generate signatures from emerging malware campaigns
Establish leakage-free evaluation and regression testing frameworks
Optimize inference throughput, latency, memory, and cost
Requirements: At least five years building production machine learning systems
At least two years in an adversarial domain
Transformer training, tokenization, fine-tuning, distillation, and quantization experience
Strong Python, PyTorch, Hugging Face, and SQL skills
English proficiency at B2 level or higher
Apply senior machine learning expertise to malware classification, transformer development, evaluation discipline, and effective production inference.