Vacant job

Do you want to help shape the future of AI in Sweden?

Artificial intelligence is rapidly transforming society, and AI Sweden is leading the development by accelerating the use of AI for the benefit of society, competitiveness, and everyone living in Sweden. We drive initiatives in healthcare, energy, and the public sector while exploring new innovations in natural language processing and edge learning. We are now looking for an MLOps specialist for our groundbreaking project, Digital Assistant Svea - a generative AI solution developed for the Swedish public sector.

About the project and the role

Digital Assistant Svea is a national initiative that brings together over 50 public organizations to create a common AI solution within the public sector. We work holistically with technology, legislation, data, and education to build a solution that delivers real value for employees in the public sector.

As an MLOps specialist, you will be part of a cross-functional team of developers and data scientists. You will play a key role in optimizing and scaling our infrastructure for language models and RAG systems (Retrieval Augmented Generation) to ensure a stable, efficient, and high-performance platform.

We primarily seek someone with solid experience, but we are also open to candidates with less experience but great potential. For the right person, there is a unique opportunity to grow with the task and take responsibility for critical parts of our technical infrastructure. Regardless of your current level of experience, the project offers a chance to develop advanced MLOps skills in an initiative with real societal impact while working with the latest AI technologies.

Your main responsibilities

• Optimize and scale infrastructure for large-scale language models across multiple GPU servers
• Develop efficient interfaces between hardware and software for AI models
• Design and improve RAG pipelines for different query types, including specialized retrieval with function calling
• Create solutions for managing large document volumes in vector databases
• Optimize performance for embedding models, rerankers, and other components in the AI system
• Conduct performance testing and benchmarking on various hardware platforms
• Implement monitoring and measurement systems for key performance metrics
• Develop intelligent queuing systems for efficient resource management
• Continuously improve system architecture to meet growing needs
• Document technical solutions and share knowledge within the team

We believe you have

• Experience working with large language models in production environments
• Knowledge of optimizing AI inference on GPU-based systems
• Experience with RAG systems and vector databases
• Strong programming skills in Python
• Understanding of Linux and containerization
• Knowledge of performance optimization for AI models and data flows
• Experience in system design for handling high loads
• Familiarity with working in teams using agile methods

It is a plus if you have previous experience with

• Experience working with open-source models
• Knowledge of embedding models and inference optimization
• Experience with object and document storage solutions
• Understanding of RAG strategies and optimization
• Experience in prompt engineering

Our technical environment

• We operate in a dynamic and innovative technical environment with the following technologies:
• Large language models: Llama 3.3 70B via vLLM
• RAG technology: Embedding models via TEI, vector database with Qdrant, reranking with BGE models
• Backend infrastructure: Postgres for document management, Minio for object storage
• Hardware platform: NVIDIA H100, L40s

Why work for AI Sweden?

This is more than just a job – it is an opportunity to be the driving force behind a groundbreaking project. You will have the chance to work with the latest technology, lead a talented team, and make a real difference for the future of digital assistant technology. If you are ready to challenge yourself and take your career to the next level, get in touch.

Step into a role where your skills, vision, and leadership will shape the future of AI. Join our team and help take Digital Assistant Svea to new heights! For us, artificial intelligence is not just about technology; it is a force for positive societal change. You will work alongside leading AI experts, researchers, journalists, linguists, policy experts, entrepreneurs, change leaders, and many more. To work here, you do not need to know "everything" about AI, but you must believe in its potential to shape our society for the better.

As an organization, we are uniquely positioned at the intersection of governmental influence and startup agility. We are small enough to be adaptable and have fun, but we maintain close contact with both the business sector, academia, and the private and public sectors.

With us, you will make a real difference by contributing to initiatives that benefit society and address critical challenges. Be at the forefront of AI innovation, work with the latest technology, and play a key role in shaping the future of AI in Sweden.

In short, we believe we offer our team members a place to grow and an environment for personal development.

An equal and fair work environment

We strongly believe in diversity and inclusion and actively strive to assemble a diverse team in terms of age, gender, and background.

We have collective agreements and offer everything included in this, such as vacation, pension, wellness benefits, and more.

If you have any questions or concerns, please do not hesitate to contact Viktoria Stark at viktoria.stark@ai.se

AI Sweden has chosen not to collaborate with external parties for this recruitment and kindly asks not to be contacted by any advertising agencies, recruitment agencies, or staffing companies.