Search fundamentals.
Study indexing, retrieval, relevance and how computational systems locate information.
Membranch is an independent technology research company exploring the foundations of information retrieval, software systems and AI evaluation. Today, our work is research—not product development.
Identify what needs to be understood.
Isolate variables and examine behavior.
Record results, limits and open questions.
Our current phase is foundational research, not a commercial offering.
We start with technical questions rather than a product brief. Our interests span search fundamentals, the behavior of complex software systems, and the methods we use to evaluate them.
Study indexing, retrieval, relevance and how computational systems locate information.
Examine performance, modularity and the practical limits of different technical approaches.
Use repeatable experiments to distinguish promising hypotheses from measured improvements.
Membranch is in an early research phase. We design small experiments, compare them with established techniques and document the limits of our findings. Our conclusions follow the evidence, including when results are inconclusive.
Form a clear, testable question and decide what would count as meaningful evidence.
Use small, focused prototypes to isolate behavior, test alternatives and understand trade-offs.
Compare results, resource use and failure cases. Share findings when they are ready to be understood in context.
ENQIX enables AI agents to search, navigate, and retrieve relevant context on demand using indexed search and agent-directed exploration.
How we research