About the role

Senior Machine Learning Engineer designing and optimizing machine learning systems for insurance and investment solutions at Equisoft. Focusing on large language model fine-tuning and synthetic data generation initiatives.

Responsibilities

Lead fine-tuning initiatives for open-source LLMs (such as Llama, Mistral, and domain-specific models) to optimize performance for insurance and financial services use cases
Design and implement scalable synthetic data generation pipelines using advanced techniques including GANs, VAEs, and LLM-based generation methods
Create and curate insurance-specific datasets for model training, ensuring compliance with privacy regulations and industry standards
Optimize machine learning models for cost-effective inference, implementing techniques such as model distillation, quantization, and efficient deployment strategies
Collaborate with the cloud developer (our ML architecture team) on overall system design, model integration, and performance optimization
Establish and maintain best practices for model versioning, deployment, and monitoring using MLOps frameworks
Develop and maintain automated model evaluation pipelines to ensure consistent performance and quality
Research and implement state-of-the-art ML techniques including transfer learning, few-shot learning, and domain adaptation
Monitor model performance in production and implement continuous improvement strategies
Work cross-functionally with product and engineering teams to integrate ML models into customer-facing applications

Technical Bachelor's Degree in Computer Science, Machine Learning, Data Science, or Engineering, or College Diploma combined with 5+ years of relevant experience
Minimum of 4 years' experience in machine learning engineering, with demonstrated expertise in model development, deployment, and optimization
Extensive experience with Python and ML frameworks including PyTorch, TensorFlow, Hugging Face Transformers, and scikit-learn
Proven experience in fine-tuning large language models (LLMs) for domain-specific applications
Strong background in synthetic data generation techniques and data augmentation methods
Experience with cloud platforms (AWS, Azure, or Google Cloud) and their ML services
Proficiency in MLOps tools and practices including model versioning, deployment pipelines, and monitoring
Knowledge of distributed computing frameworks (Apache Spark, Dask) for large-scale data processing
Experience with containerization technologies (Docker, Kubernetes) for model deployment
Understanding of statistical modeling, deep learning architectures, and optimization techniques
Excellent knowledge of French & English (spoken and written)