Applied Scientist (Time Series)

Applied Scientist Time Series

Company

Mistral AI

Conditions

5 days ago

Skills

Candidate Availability

Required and preferred rules are kept separate and reflect the wording in the original posting.

About the Role

You will train and deploy time-series foundation models at scale. You will generate datasets, build evaluators and benchmarks, advance model architectures and training methods, and adapt models to client data. You will develop supporting tools, collaborate on complex use cases, and manage research projects and client communications.

Requirements

• Fluency in English
• PyTorch or JAX expertise
• Knowledge of deep learning and large language models
• Experience pre-training or fine-tuning time-series foundation models or large transformer models at scale
• Time-series modelling expertise including probabilistic forecasting, multivariate series, covariates, statistical models, and gradient boosting
• Ability to contribute independently to a large codebase
• Python programming
• Track record through personal projects, professional projects, or academia

Responsibilities

• Run pre-training, post-training, and deployment of time-series foundation models
• Generate and curate large-scale real and synthetic time-series datasets
• Build evaluators and benchmarks for forecasting and time-series tasks
• Advance architectures, training recipes, scaling, and model capabilities
• Fine-tune and adapt models to client data
• Benchmark models against existing client models
• Develop tools and frameworks for data generation, training, evaluation, and deployment
• Collaborate on complex use cases involving agents and language models
• Manage research projects and client research communications

Benefits

• Healthcare coverage
• Parental leave
• Retirement plans
• Relocation support
• Wellness programs
• Meal allowances
• Transportation allowances
• Location-specific perks

Показать контакты