ml engineer for time-series foundation models

Mistral

Paris

Presencial

EUR 120.000 - 180.000

Tempo integral

Há 11 dias

Uma candidatura completa em um minuto

Currículo e carta de apresentação personalizados, prontos para serem enviados.

Passe pelos filtros ATS

Benefícios oferecidos por esta vaga

Медицинское страхование
Компенсация переезда
Программы оздоровления
Питание и транспорт

Resumo da vaga

Компания Mistral ищет специалиста по временным сериям для запуска предобучения, постобучения и развёртывания в масштабах тысячи GPU. Ваша работа будет включать создание и оценку наборов данных реальных и синтетических временных рядов, а также разработку инструментов для обучения и развёртывания.

Требуются глубокие знания PyTorch/JAX, опыт работы с LLM и временными рядами, способность работать автономно и эффективно общаться с клиентами.

Qualificações

  • Свободное владение английским языком и способность объяснять сложные концепты различным аудиториям.
  • Экспертные знания PyTorch или JAX
  • Твёрдое понимание современных DL и LLM, включая предобучение, постобучение и оценку
  • Опыт предобучения или тонкой настройки временных рядов или крупных трансформеров в масштабе

Responsabilidades

  • Запуск предобучения, постобучения и развёртывания современных временных рядов на кластерах с тысячами GPU
  • Генерация и куруирование наборов данных, включая синтетические данные
  • Улучшение архитектур Timestral, рецептов обучения и масштабирования
  • Тонкая настройка Timestral на клиента и сравнение с их моделями
  • Разработка инструментов и фреймворков для генерации данных, обучения, оценки и развёртывания
  • Сотрудничество с кросс-функциональными командами по сложным кейсам, включая агентов, работающих с LLM
  • Управление исследовательскими проектами и коммуникациями с клиентскими исследовательскими командами

Conhecimentos

English fluency
PyTorch
JAX
Time-series modelling
LLM knowledge

Formação

PhD in ML / Stats / Applied Math / Physics

Descrição da vaga

Описание:

Mistral provides full-stack AI solutions, from frontier models to developer tools, applications, and compute. It partners with enterprises in high-stakes industries to co-create customized AI systems that can be run on their terms.

Задачи:
  • Run pre-training, post-training, and deployment of state-of-the-art time-series foundation models on clusters with thousands of GPUs
  • Generate and curate large-scale real and synthetic time-series datasets, and build evaluators and benchmarks for forecasting and other time-series tasks
  • Advance Timestral architectures, training recipes, scaling, and capabilities such as anomaly detection, classification, and embeddings
  • Fine-tune and adapt Timestral on client data in the cloud or on-premise, and benchmark it against clients' existing models
  • Develop tools and frameworks for data generation, model training, evaluation, and deployment
  • Collaborate with cross-functional teams on complex use cases, including agents combining Timestral with Mistral LLMs
  • Manage research projects and communications with client research teams
Требования:
  • Fluent English and excellent communication skills; able to explain complex technical concepts to technical and non-technical audiences
  • Expert knowledge of PyTorch or JAX
  • Solid knowledge of modern deep learning and large language models, including architectures, pre-training, post-training, and evaluation
  • Experience pre-training or fine-tuning time-series foundation models or large transformer models at scale
  • Strong background in time-series modelling, including probabilistic forecasting, multivariate series, covariates, and classical baselines such as statistical models and gradient boosting
  • Able to contribute to a large codebase and work independently with little guidance
  • Writes clean, readable, high-performance, fault-tolerant Python code
  • Works independently without roadmaps and is comfortable with uncertainty
  • Collaborative, low-ego, and eager to learn
  • Track record of success through personal projects, professional projects, or academia
Будет плюсом:
  • founder or builder experience
  • a PhD or Master's in Machine Learning, Statistics, Applied Mathematics, or Physics
  • research on time-series foundation models
  • hands-on experience training or fine-tuning LLMs
  • building agents and tool use, or models combining text and time series
  • industry experience in energy, industrial operations, telecom, or finance
  • experience deploying models for industrial clients
  • experience with large codebases or large machine learning models at scale
  • contributions to a large open-source or industry codebase
  • publications in top academic journals or conferences
  • experience improving existing code through typing, tests, and CI pipelines
Условия:
  • healthcare coverage
  • parental leave
  • retirement plans
  • relocation support
  • wellness programs
  • meal and transportation allowances
  • other location-specific perks
Obtenha sua revisão de currículo grátis e confidencial.

ou arraste e solte seu arquivo aqui.

Similar jobs

Vagas similares que valem a pena comparar

ml engineer for operational research
ml engineer for operational research

Mistral • Paris

Presencial
EUR 90.000 - 150.000
Healthcare
Parental leave
Retirement plans
+4
ml engineer for formal methods
ml engineer for formal methods

Mistral • Paris

Presencial
EUR 90.000 - 130.000
Healthcare coverage
Parental leave
Retirement plans
+2
ml engineer for tabular data
ml engineer for tabular data

Mistral • Paris

Presencial
EUR 90.000 - 150.000
Healthcare coverage
Relocation support
Meal allowances
Applied Science - Time Series
Applied Science - Time Series

Mistral • Paris

Híbrido
EUR 90.000 - 130.000
ml team lead in robotics
ml team lead in robotics

Mistral • Paris

Presencial
EUR 120.000 - 190.000
Healthcare coverage
Relocation support
Meal and transport allowances
Applied Science - Time Series
Applied Science - Time Series

Mistral.ai • Paris

Presencial
EUR 90.000 - 150.000
ml engineer in robotics
ml engineer in robotics

Mistral • Paris

Presencial
EUR 90.000 - 160.000
Healthcare coverage
Parental leave
Retirement plans
+3
Evaluation Engineer, Applied AI
Evaluation Engineer, Applied AI

Mistral • Paris

Presencial
EUR 70.000 - 110.000
Benefits package
Forward Software Engineer, Data
Forward Software Engineer, Data

Mistral • Paris

Presencial
EUR 90.000 - 150.000
Healthcare coverage
Relocation support
Wellness programs
+1
Applied AI, Frontier Data Engineering
Applied AI, Frontier Data Engineering

Mistral • Paris

Presencial
EUR 90.000 - 140.000
Healthcare coverage
Relocation support
Well-being programs