Вакансия из Hirify Global, списка международных tech-компаний Для мэтча и отклика нужен Plus
Пожаловаться
Зарплата и рынок
ЗП не указанаразвернуть
На похожих ролях по рынку
$46к$73кмедиана$144к
12 в выборке
На СНГ рынке: $2.7к/мес ($1.6к - $3к)
91
Крутая вакансия
развернуть
Чётко описанная роль в сфере AI у ведущей аэрокосмической компании даёт возможность работать над ML-системами для реальных продуктов, хотя два направления требуют разного опыта.
Современный стек
Понятные задачи
AI в авиации
Развитие внутри группы
Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Подключить Plus
Описание вакансии
Текст:
Hirify AI/Оригинал
TL;DR
ML Data & Infrastructure Engineer (AI): Building engineering datasets and high-performance ML training infrastructure on sovereign GPU clusters with an accent on data pipelines, synthetic data, reproducibility, and secure on-premise systems. Focus on operating distributed GPU workloads, governing data provenance and compliance, and enabling reliable model training, tracking, and release.
Location: Munich, Madrid, Toulouse, Paris, or Hamburg
Company
hirify.global develops aerospace, defense, space, and connected-service technologies, with a focus on sustainable and technologically advanced aircraft and security solutions.
What you will do
Build modular, documented, tested, and secure codebases for ML engineers and applied scientists in an on-premise environment handling sensitive aerospace data.
Architect ingestion pipelines for heterogeneous engineering data, including documents, requirements, reports, images, and 3D data; clean, deduplicate, version, and prepare datasets for training and evaluation.
Generate synthetic engineering data using foundation models and physical rules, and establish labelling and data-quality processes with engineering experts.
Track data provenance, licensing, and export-control restrictions such as ITAR/EAR to keep datasets auditable and production-ready.
Operate GPU-cluster training environments, including scheduling, storage, resource sharing, experiment tracking, model registries, and release pipelines.
Monitor GPU utilization, memory bottlenecks, token and compute costs, capacity forecasts, and interfaces for model serving and security.
Requirements
Master’s degree in computer science, data engineering, or a related field.
Several years of hands-on data engineering or ML infrastructure experience in an industrial or product setting.
Strong Python skills and solid software engineering practices, including testing, version control, and CI/CD.
Fluent English required.
Training data track: experience building large-scale data pipelines with technologies such as Spark, Ray, or Dask and preparing data for ML training.
Training data track: experience extracting and structuring unstructured data such as PDFs, technical documents, or images.
ML infrastructure track: experience running ML workloads on GPU clusters with Slurm or Kubernetes and containers.
ML infrastructure track: experience with MLOps tools such as MLflow, Weights & Biases, or similar.
Nice to have
Experience generating synthetic data with LLMs or curating data for LLM training.
Experience with labelling tools and human-in-the-loop data processes.
Experience in air-gapped or security-sensitive environments.
Infrastructure as code with Terraform or Ansible and GPU monitoring experience.
Background in engineering, manufacturing, or aerospace data.
German language skills.
Culture & Benefits
Permanent, full-time position within hirify.global.
Work with open models, sovereign on-premise infrastructure, and real engineering data for aircraft development.
Opportunity to influence AI use in aircraft engineering and products.
Professional development opportunities across the hirify.global group.
Flexible collaboration arrangements are supported where possible, alongside a startup-oriented team environment within an established technology group.
Будьте осторожны
Если работодатель просит войти в их систему, используя iCloud/Google, прислать код/пароль, запустить код/ПО, не делайте этого. Это мошенники. Обязательно жмите «Пожаловаться» или пишите в поддержку. Подробнее в гайде →