AI Engineer/Language Technologist

AI Engineer/Language Technologist – Language Models Section

Do you want to work on developing Norwegian language models to understand Norwegian language and culture? Are you interested in machine learning, artificial intelligence and large language models? Are you concerned that machines should speak Norwegian, understand Norwegian dialects and know Norwegian conditions? Then this is something for you.

Over the past twenty years, the National Library has established one of the world's largest collections of digital cultural heritage, including digital text. In 2024, we launched NB-Whisper, the best model for speech recognition in Norwegian to date. We also carried out an exciting project to train and evaluate a number of large generative text models based on digital text from our collection.

The National Library has been commissioned by the government to train and make available Norwegian and Sámi language models for use in the public and private sectors from 2025. To achieve this, we will further develop our AI lab into a separate section for language models. Would you like to help us strengthen one of Norway's leading environments for the development of language models and AI at the National Library?

The National Library of Norway is now advertising one to two permanent positions and up to three 2-year engagements as AI Engineer/Language Technologist. We need people who will help organize and further develop our digital collection for the development of language models for Norwegian and Sámi. We need those with expertise in processing large amounts of text or speech data, or who have experience in training and evaluating language models. We are interested in the full cycle of language model development: data preparation and selection, pre-training, fine-tuning and adaptation, reasoning, and evaluation.

The National Library has a close collaboration with Sigma2, which provides the necessary computing power for our work. The National Library is also one of the main partners in a AI center application for language models to the Research Council of Norway.

Job duties

• Preparation of datasets for training and evaluating models
• Training and evaluation of language models
• Fine-tuning models for specific domains/tasks
• Instructional adjustment, adaptation, and reasoning
• Making models available and user support
• Documentation of results and contributions to scientific publications

Qualifications

We are looking for someone with an education equivalent to at least a master's degree in computer science/informatics, preferably with a specialization in language technology or machine learning, or equivalent work experience with a relevant educational background, e.g. linguists/digital humanists who have ventured into NLP. Both newly graduated and experienced developers are welcome! We emphasize collaboration, competence development and knowledge sharing.

You must be willing to learn and have interest, understanding and knowledge of several of the following areas:

• Development, evaluation and use of machine learning models
• Development of language technology and data management in e.g. Python
• Kubernetes, Docker, Git, Slurm or Spark
• It is an advantage if you are fluent in Norwegian as a working language, both spoken and written, but this is not a requirement. We also need people with knowledge of Sámi languages.
• Experience from similar work is desirable.

Personal characteristics

• Structured and accurate
• Good collaboration skills
• Be able to communicate clearly and concisely about language technology issues
• Ability to work independently and effectively
• Personal suitability is emphasized.

Number of positions: 1-2 permanent positions and up to 3 temporary positions

• Permanent or temporary position in position code 1087 (chief engineer), 1181 (senior engineer) or 1109 (researcher II). Paid according to the state's salary regulations in the salary range of 650,000 - 800,000, depending on qualifications, expertise and experience. For particularly qualified applicants, higher salaries may be considered.
• Membership in the State Pension Fund.
• The National Library is an IA enterprise.

Work location

Positions are located in Mo i Rana or Oslo, or by agreement, with a preference for Mo i Rana.

About the National Library of Norway

The National Library is one of the country's largest and most important cultural institutions, and one of the foremost sources of knowledge about Norway and Norwegian affairs. Based on the Legal Deposit Act, we collect and preserve what has been published in Norway in all media types, such as books, magazines, newspapers, music, film, photography, in broadcasting and online. By virtue of the collection and by making it available via digital services, the National Library forms the basis for documentation, research and learning, and contributes to creating identity and belonging.

The National Library will contribute to the further development of Norwegian libraries as active and relevant social institutions, and aims to be one of Europe's most exciting and modern national libraries.

We have approximately 600 employees in Oslo and Mo i Rana, and a budget of NOK 845 million.

The National Library of Norway wishes to reflect the diversity of the population and participates in the government's inclusion efforts. We therefore encourage qualified candidates to apply, regardless of age, gender, disability, ethnic background, sexual orientation or gaps in their CV. Working conditions will be facilitated for employees with disabilities.

Information about disabilities or gaps in the CV may be used for registration purposes. The names of all applicants may be published even if the applicant has requested not to be included on the official applicant list, cf. the Public Access Act, Section 25. Copies of certificates and diplomas must be attached to the application, and originals must be presented at the interview.

Post in FINN.no: https://www.finn.no/job/fulltime/ad.html?finnkode=402418276