TL;DR
Member of Technical Staff, AI Data (AI): Designing and developing data pipelines for the world’s most advanced multimodal dataset with an accent on large-scale data ingestion, processing, and infrastructure. Focus on rigorous experimentation, identifying model gaps, and optimizing data recipes to power cutting-edge AI frontier models.
Location: Onsite in London, United Kingdom
Company
hirify.global is building the largest and most advanced multimodal dataset in the world to power the training of capable AI frontier models, driving the consumer Copilot experience.
What you will do
- Design and develop data pipelines for ingesting enormous amounts of multi-modal training data (text, audio, images, video).
- Build and maintain cutting-edge infrastructure to store and process petabytes of data needed to power AI models.
- Partner with pretraining and post-training teams to improve data recipes through rigorous and careful experimentation.
- Collaborate with product teams, engineers, and researchers across hirify.global to identify gaps in current generation models.
Requirements
- Bachelor’s Degree in Computer Science, Math, Software Engineering, Computer Engineering, or a related field, AND experience in business analytics, data science, software development, data modelling, or data engineering work, OR equivalent experience.
- Expertise in large-scale data engineering, ideally applied to AI.
- Expertise in Spark, Kubernetes, or similar technologies.
Culture & Benefits
- Opportunity to contribute to the next generation of systems that will transform the field of AI.
- Work in a highly collaborative, fast-paced, and interdisciplinary team environment.
- Emphasis on craftsmanship, attention to detail, proactive attitude, and exploring new methods and technologies.
- Microsoft is an equal opportunity employer committed to diversity and inclusion.
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 →