Origin & Lineage
Creator and Research Lineage
NINMENI does not come from a university lab, a funded startup, or a corporate AI research group. Understanding where it comes from is part of understanding what it is.
The Researcher
NINMENI is created and maintained by Emylton Leunufna (also known as Paulus Femi Leunufna), an independent researcher from Indonesia. The path is self-taught — no formal computer science degree, no AI lab affiliation, no institutional mentorship.
This is not framed as a disadvantage. The absence of institutional framing means the research is not constrained by publication pressure, grant requirements, or the need to compare favorably against existing paradigms. The question can be asked on its own terms.
Intellectual Lineage
The ideas behind NINMENI draw from several traditions that are underrepresented in mainstream deep learning research:
- Cognitive science and the study of how meaning is actually formed in biological systems.
- Formal semantics and the tradition of treating meaning as compositional structure.
- Philosophy of mind and questions about what genuine understanding requires.
- Computational linguistics before the scaling era — when structure mattered more than data.
These traditions share a belief that meaning is not a by-product of statistical regularities in large corpora. NINMENI is built on that belief.
Why Indonesia?
Indonesia-native AI research is rare. Most AI systems are built on English-language corpora and retrofitted to other languages. NINMENI is built from the inside of Indonesian — treating it not as a target language to be handled but as the native environment of the system.
This is not a nationalistic claim. It is an architectural one: a system built natively for a language handles that language structurally, not as a second-class approximation.
Governance Note
This project is independent. There is no board, no investors, no external review committee, and no timeline pressure imposed by outside parties. Decisions about what to build, what to publish, and what to claim are made by the researcher alone, governed only by a rigorous internal epistemic standard. Progress is measured by what the architecture teaches, not by what external validators expect.