Our Mission

We aim to provide a flexible and transparent framework for the application of large language models (LLMs) in the realm of biomedical research and applications. We put emphasis on an open-source process, which is vital to keep up with the rapid pace of LLM development. This means we take great care to not unnecessarily reimplement solutions that exist in the broader field, such as the model chaining workflows provided by LangChain.

Rather, we look to provide an interface tuned towards the needs of the biomedical community, with focus points on robustness in the face of health-critical applications, native connectivity to data representations used in biomedical research, and a benchmark-forward approach to model evaluation in very specific use cases.

Our Vision

Our developments have the overarching goal of making biomedical research more productive by improving accessibility of technical workflows. Importantly, we aim to automate low-level tasks to allow researchers to spend more time on higher-level reasoning and data analysis without being burdened by the tedious aspects of handling data and knowledge.

Inspired by the supportive role of Integrated Development Environments (IDEs) in programming, we want to support research applications with an "Integrated Research Environment" (IRE) in the longer term. The development of this system cannot be done in isolation; if you are interested in contributing in any way, please get in touch!