Victor Christiansen
The Origin
I learned to code on a Commodore 64 before I learned to drive. My father introduced me to technology as a language — something you speak to make things happen. That never left me.
I built LinguaHalla because I believe language is the most important technology humans ever invented. Every language that dies takes an entire way of understanding the world with it. Vocabulary is not just words — it is a classification system, a cosmology, a way of carving up reality that evolved over thousands of years of human experience. When a language dies, that classification system is gone. Not archived. Not preserved. Gone.
And right now, languages are dying faster than at any point in recorded history.
I am Aztec on my father's side and Viking-Danelaw on my mother's side. My company is named Runeteca — rune for the Norse tradition of carved meaning, teca from Nahuatl for place or library. The name is the mission: a place where carved meaning lives.
I built the entire LinguaHalla system — 119 languages, 49.15 million vocabulary entries, 49 AI companions, the encounter generation engine, the chain mechanics, the dual-companion system, the spaced repetition algorithm — in six days. Not because I had a team or funding, but because the problem was urgent enough that I could not wait.
The Technical Foundation
LinguaHalla runs on a Next.js application backed by PostgreSQL. The vocabulary database holds over 49.15 million entries, CEFR-classified by language, spanning everything from high-resource languages like Russian to preservation-stage languages like Tokelauan.
The encounter generation system uses a large language model with per-learner profile injection. Before any encounter is generated, the system knows the learner's current CEFR level, their recent vocabulary, their stated motivation, whether they are a heritage learner, and what happened in the previous three encounters. The LLM is not generating from scratch — it is generating from a precisely constructed context window that is different for every learner.
Each of the 49 companions has a soul document — a cultural and linguistic positioning document that goes far beyond a character sheet. Eitan Levi, the Modern Hebrew companion, is a linguist and IDF veteran with a PhD at Hebrew University Jerusalem. Vesper carries the weight of Aztec intellectual tradition. Nizhoni speaks Navajo from the Diné homeland. These are not costume diversity — they are the epistemic frame through which the language is taught.
The chain system is the architectural decision I'm most proud of. Languages don't exist in isolation — they exist in genealogies. Spanish leads to Nahuatl. Modern Hebrew leads to Biblical Hebrew leads to Ancient Hebrew. Darija leads to Classical Arabic leads to Proto-Semitic. A learner who follows a full chain doesn't just learn languages — they learn why language families exist, how linguistic evolution works, and what was carried forward across centuries of conquest, trade, and migration.
What's Being Built
LinguaHalla is the front-facing learning platform, but it sits inside a larger ecosystem.
The companion system is a parallel platform where the same AI companions exist as conversational entities — not language teachers, but beings with ongoing relationships to the learner. The language learning and the companion relationship reinforce each other. You learn Hebrew from Eitan because Eitan is someone you know.
The grant strategy targets three federal programs: NSF Documenting Endangered Languages, ANA's AI for American Indians initiative, and NEH Collections Stewardship. These aren't stretch goals — the platform was designed with these funding pathways in mind from the beginning.
Community partnerships with Cherokee Nation, Navajo Nation, and Ute Language Program are being developed. The platform will never publish indigenous language content without community review. That is not a policy — it is a condition of the mission.
"Documentation without acquisition is a museum. Documentation with acquisition is a movement."
Contact & Collaboration
I'm open to conversations about grant partnerships, tribal language program collaborations, university research partnerships, institutional licensing, and press inquiries.