Requirements: English
Company: Storm3
Region: Paris , le-de-France
LLM Engineer w/ particular focus on speechprocessing and integration Hybrid in Paris Competitive base TheMission To ttackle the fundamental challenges of world modeling andestablish a new paradigm for next-generation machine reasoning.They are looking for passionate individuals who share our visionand are eager to push the boundaries of AI together. KeyResponsibilities: Data Infrastructure & Pipelines - Design,implement, and maintain scalable video data pipelines to supportlarge-scale training. - Develop data preprocessing, transformation,and synthesis workflows to support world model training. -Contribute to building high-quality data annotation pipelines toensure accurate and consistent labels across large-scale datasets.Key Responsibilities: Training & Inference Systems - Supportthe training of multimodal foundation models (e.g., video diffusionmodels, world models) by developing and optimizing distributedtraining systems. - Improve inference and serving efficiency forreal-time interaction through model optimization and system tuning.- Monitor system health and performance, and contribute todebugging and optimization at scale. Key Responsibilities:Collaboration & Integration - Work closely with research teamsto understand experimental goals and translate ideas into reliableand maintainable infrastructure and tools. - Integrate novelresearch prototypes into production-ready systems and ensurereproducibility at scale. - Participate in design and code reviews,ensuring code quality, efficiency, and compliance with bestpractices. Key Responsibilities: Benchmarking & Evaluation -Contribute to the development of tools and infrastructure toevaluate model performance using rigorous quantitative benchmarks,including metrics for physical accuracy and controllability. KeyResponsibilities: Codebase & Documentation - Maintain andextend shared codebases, contribute to internal documentation, andsupport onboarding of new team members or collaborators. - Writeclean, efficient, and well-tested code for components across themodel development lifecycle. Key Responsibilities - Supportcontributions to research papers and demos when engineering workplays a significant role. - Help represent the teams engineeringexcellence in internal and external forums when appropriate.Academic Qualifications - MSc or PhD in Machine Learning orComputer Science, or equivalent industry experience. ProfessionalExperience Required - Proficient in data collection, cleaning, andtransformation at scale, including designing robust pipelines formultimodal datasets (e.g., video, audio, text). - Practicalexperience with web scraping and crawling frameworks (e.g., scrapy,selenium, playwright, BeautifulSoup) to collect and curatehigh-quality web-scale datasets. - Experience in large-scale modeltraining (LLMs or Diffusion Models) on large clusters. - Hands-onexperience with state-of-the-art video generative models (e.g.,Sora, Veo2, MovieGen, CogVideoX, etc.). - Experiences in buildingand optimizing large-scale video data pipelines. - Experience inaccelerating diffusion model inference for improved efficiency. -Exceptional problem-solving and troubleshooting skills to tacklecomplex technical challenges. - Strong systems and engineeringexpertise in deep learning frameworks such as PyTorch. - Strongcommunication and collaboration skills for effectivecross-functional teamwork. - Demonstrated ability to solve complexsystem-level challenges and debug failures across thetraining/inference stack (e.g., memory issues, deadlocks, I/Obottlenecks).