Research Engineer, Applied AI

Kapa makes technical knowledge instantly accessible through AI assistants. As a research engineer you will work on improving Kapa’s ability to answer harder and harder technical questions. Check out Docker’s documentation ([https://docs.docker.com](https://docs.docker.com/)) for a live example of what Kapa is (look for the “Ask AI” button). **The following challenges should excite you:** * Evaluating a RAG system in production without labelled data. * Creating your own benchmark from scratch. * Building an agentic retrieval system that can judge when to be fast and when to take more time. * Fine tuning embeddings or rerank models. **To solve these challenges you will:** * Work directly with the founding team and our software engineers. * Keep up with the latest developments in the space and see how they can be applied. * Design and run experiments. **You may be a good fit if

you have:

** * A Master's/ PhD degree in Computer Science, Machine Learning, Mathematics, Statistics or a related field. * A detailed understanding of machine learning, deep learning (including LLMs) and natural language processing. * Hands-on experience in training, fine-tuning and deploying large language models. * Have prior experience working with vector databases, search indices, or other data stores for search and retrieval use cases. * Significant experience building evaluation systems for LLMs or search. * Familiarity with various information retrieval techniques, such as lexical search and dense vector search. * The ability to work effectively in a fast in a environment where things are sometimes loosely defined. * Want to learn more about machine learning research. \* This is neither an exhaustive nor necessary set of attributes.

Even if none of these apply to you, but you believe you will contribute to kapa.ai, please reach out. **Location**: Remote within Europe. We’re a distributed team and welcome applicants based anywhere in Europe. We also have an office in Copenhagen for those who prefer working on-site or hybrid.