Description
Ericsson is seeking a researcher to study how AI code knowledge bases affect developer productivity and answer quality across the software development lifecycle. The role involves designing evaluation methodologies, implementing telemetry and instrumentation, running controlled studies comparing knowledge-base-assisted and baseline AI coding assistants, analyzing results, and delivering a reusable evaluation framework and written study. Candidates should be pursuing a Master's degree in a relevant field, have solid Python and LLM/agent tooling skills, and be interested in empirical software engineering and AI system evaluation.
