Junior AI Evaluation

Há 16 horas

funchal, ilha da madeira, Portugal Associação Fraunhofer Portugal Research Tempo integral

Organisation/Company Associação Fraunhofer Portugal Research Department Human resources Research Field Computer science » Programming Engineering » Biomedical engineering Computer science » Other Engineering » Computer engineering Computer science » Informatics Researcher Profile First Stage Researcher (R1) Positions Master Positions Application Deadline 31 Oct 2026 - 23:59 (Europe/Lisbon) Country Portugal Type of Contract Other Type of Contract Extra Information Other Job Status Full-time Hours Per Week 40 Offer Starting Date 1 Dec 2026 Is the job funded through the EU Research Framework Programme? Not funded by a EU programme Reference Number AICOS_Jobs_2026_04 Is the Job related to staff position within a Research Infrastructure? Yes

Offer Description

Junior AI Evaluation / Trustworthy AIScientist

Job Title: JuniorAI Evaluation / Trustworthy AI Scientist

As a researcher at AICOS you will work at the forefront of AI innovation, creating purposeful technologies that positively transform industries, promoting social well-being and improving quality of life.

Specifically, in this job you will participate in the research and development of components for an evaluation framework to apply in the context of LLMs, agentic AI and physiological time-series data, namely in the dimensions of robustness and uncertainty, and bias/fairness, towards transparent and responsible AI practices.

Your role:

  • Research and develop Responsible AI methodologies related to the evaluation of LLM-based and agentic AI systems, namely:
    • Investigate robustness, uncertainty, and reliability, including sensitivity to input variations, distribution shifts, and potential failure modes;
    • Develop and apply methodologies to assess bias, fairness, and performance variability across relevant population subgroups;
    • Design and implement reproducible evaluation pipelines, systematically analyzing model and system behavior, limitations, and potential risks;
  • Provide methodological guidance for evaluating AI models based on physiological time-series data, addressing aspects such as data quality, robustness, uncertainty, and performance variability across