Suggests, drafts and organises
Only where you choose to use it, and always shown with the evidence it is based on.
About Byleron
Built from research. Reimagined for researchers.
We didn’t begin with a technology. We began with the experience of doing research.
Question
Evidence
ByleronResearch intelligence
Interpretation
Knowledge
Why Byleron exists
We found ourselves adjusting our work to fit the tools.
Byleron grew from years of working with academic research tools, discovering their capabilities, and encountering their limitations.
Existing software helped us accomplish important tasks. But as our research became more complex, we repeatedly encountered a familiar problem: tools that were powerful in individual functions did not always support the way researchers actually work.
Sometimes the necessary feature was missing. Sometimes it existed, but the workflow made a straightforward research task unnecessarily complicated. And sometimes the software’s design seemed to determine how the research should proceed, rather than adapting to the researcher’s methodological needs.
We began asking a different question: what if research software were designed around the researcher, rather than requiring researchers to work around the software?
The research workflow
Rethinking how research fits together.
Research is not a collection of isolated actions. A researcher moves between questions, literature, evidence, coding, comparison, interpretation, and writing. These activities inform one another: a new finding can lead back to an earlier source, a comparison can challenge an emerging interpretation, and a change in the analytical framework can require returning to the original data.
Yet research software often divides these connected activities into separate functions, interfaces, or products. We wanted to reconsider those boundaries.
Instead of starting with a catalogue of features, we started with the research process itself: what researchers need to accomplish, how decisions develop, and where technology can genuinely make the work better. That perspective became the foundation of Byleron.
Artificial intelligence
AI should assist, not replace, the researcher.
As artificial intelligence became increasingly capable, it presented extraordinary opportunities for academic research. It also raised difficult questions about accuracy, transparency, accountability, authorship, and the role of human judgement.
We did not want to introduce AI simply because it could generate codes, summaries, or interpretations. We wanted to consider whether it should perform a particular task, what evidence it could reliably use, and how researchers could evaluate its contributions.
AI can assist the research process. It should not take ownership of the researcher’s decisions.
In Byleron, we aim to make AI assistance purposeful, optional, evidence-connected, and reviewable. Researchers should be able to examine suggestions, question outputs, accept or reject proposals, and remain responsible for their interpretations.
We believe responsible AI integration is not simply a matter of adding an ethical statement to a product. It should influence how the product works.
Review, question, accept or reject
Interpretation, judgement and responsibility stay with you.
From an idea to a growing set of tools
One way of working, applied with care.
Byleron QDA brings this philosophy into qualitative research. It connects different forms of evidence, coding, analytical exploration, and results preparation within a research workspace. Its design reflects a simple principle: researchers should be able to move from their original material to their findings without losing the context and decisions that connect them.
The idea behind Byleron extends beyond qualitative data analysis. More Byleron tools are on the way, built the same way: understanding the research task first, questioning the limitations of existing workflows, and developing tools that respond to how scholarly work is actually carried out.
Our ambition is not merely to bring more tools together. It is to make research technology more coherent, capable, and responsive to the people using it.
Our principles
What guides the work.
Research comes first
Technology should adapt to scholarly needs, not dictate research methodology.
Evidence remains central
Analytical claims should stay connected to the sources and decisions that support them.
Researchers retain control
AI and automation can assist, but interpretation, judgement, and responsibility remain with the researcher.
Work stays portable
Research materials and outputs should remain usable beyond a single platform wherever practical.
Design matters
A feature is not truly useful simply because it exists. It must work in a way that makes sense for the research task.
Research deserves better tools.
What would research software look like if we designed it from the researcher’s side of the screen?
We are building our answer.