Job Responsibilities

We are looking for a research engineer who can work with uncertainties in LLM-based scientific agents.

In the project, you will (i) identify potential sources of uncertainties in AI agents, (ii) investigate ways to assess the quality of uncertainty estimates using standard methods for LLM, and (iii) develop a benchmark for uncertainty quantification in LLM-based scientific agents.

The project will be conducted under the leadership of Martin Trapp and Hossein Azizpour, and you will participate in regular group meetings. Additionally, a scientific one-day workshop with leading international researchers and partners will be planned and held during the project, where partial results will be presented. Furthermore, the results will need to be presented at the EECS Research and Impact Day in the summer of 2026.

Qualifications

Requirements

Completed degree at the undergraduate or graduate level (higher education) in the relevant field or equivalent competence.
Experience in deep learning and machine learning tools.
In-depth knowledge of LLM and Transformer architectures.
Excellent programming skills in PyTorch/JAX and experience working with GPUs and high-performance clusters.
Strong mathematical skills with an excellent understanding of statistics (e.g., hypothesis testing, Bayesian statistics).
Good collaboration skills, independence, and critical thinking.

Meritorious

In-depth experience with LLM agents.
Experience in uncertainty quantification in LLM.
Experience in Bayesian deep learning.
Experience in Deep Generative Models.
Experience in conducting large-scale experiments.
Awareness of diversity and equal treatment issues with a particular focus on gender equality.

We will place great emphasis on personal qualities.

Become a part of KTH

KTH shapes the future through education, research, and innovation. As a leading international technical university, we play an active role in driving and participating in the transition to a sustainable society. Here you are offered the opportunity to grow and develop in a creative and dynamic workplace with good working conditions and benefits. Gender equality, diversity, and equal opportunities are quality issues and an obvious part of KTH's values as a university and government agency.

More information about working at KTH (https://www.kth.se/om/jobba-pa-kth)

Union Representatives

Contact information for union representatives (https://www.kth.se/om/work-at-kth/fackrepresentanter-1.500898).

Application

You apply via KTH's recruitment system. As an applicant, you have the main responsibility for ensuring that your application is complete when submitted.

The application should include:

CV with information about relevant experience (in English).
A brief account of how your practical experience is relevant to the project and why you want to apply for this position (max 1 page, in English).

The application must be received by KTH no later than the closing date at midnight, CET/CEST (Central European Time/Central European Summer Time).

About the Employment

The employment is temporary according to agreement - for up to 6 months, with start date by agreement.

Other

For information on the processing of personal data (https://www.kth.se/om/work-at-kth/processing-of-personal-data-in-the-recruitment-process-1.823440) in connection with recruitment.

It may occur that an employment at KTH is placed in a security class. If this is the case for this position, a security review of the applicant will be conducted in accordance with the Security Protection Act (2018:585) after consent. In these cases, a prerequisite for employment is that the applicant is approved after the security review.

We kindly ask for no direct contact with staffing and recruitment companies or sellers of job advertisements.