KIANI AI
Imagine asking a question — and getting a real answer. Not bullet points. The reasoning. The trade-offs. The things a resume can't capture and a portfolio can't convey, explained by someone who knows the full story.
KAI is building that experience — an AI that represents Victor's professional identity with real depth, real context, and its own voice.
KIANIAI
KAI knows Victor's story — and speaks for itself.
Not a chatbot bolted onto a website. KAI is an agentic system — an autonomous AI that reasons about your questions, decides what information to retrieve, and composes answers with real context.
Agentic core
A wrapper around frontier LLMs that doesn't just respond — it plans. KAI decomposes questions, determines what it needs to look up, and synthesizes an answer across multiple sources.
Tool use
KAI has access to structured tools — search across professional history, retrieve case studies, cross-reference education and certifications, surface project details on demand.
Grounded knowledge
Every response is anchored to documented experience — coursework, roles, decisions, projects. The system retrieves before it generates, so nothing is fabricated.
Voice layer
A full speech pipeline — speech-to-text for input, AI voice synthesis for output. KAI doesn't just type back. It speaks.
KAI doesn't just type back — it speaks. A dedicated voice delivering every response as a spoken conversation rather than a wall of text.
Speech-to-text
Ask your question out loud. The system transcribes spoken input in real time using AI-powered speech recognition, converting natural speech into a query KAI can reason about.
Voice synthesis
KAI's response comes back spoken — not a generic text-to-speech engine, but a voice that carries the tone and cadence of an actual conversation.
Type a question or speak it aloud — about a role, a decision, a project, or anything on the portfolio. Natural language, no menus.
KAI breaks down the question, identifies what information matters, and determines which tools to call — search, retrieval, cross-referencing — before composing a response.
Structured tools pull from a knowledge base of professional history, education records, project documentation, and case studies. KAI retrieves exactly what it needs — nothing more.
Read the answer or hear KAI speak it — grounded, specific, conversational. The kind of depth you get from someone who knows the full story.
Simulated conversations showing the kind of depth KAI is being built to provide.
What made Victor leave Sweden for the U.S. at sixteen?
It wasn’t one moment — it was a pattern. Victor had been restless in Malmö for years, watching opportunities exist somewhere else. When the chance came to move to New York alone, the risk of going felt smaller to him than the risk of staying. That willingness to act on discomfort is something that shows up across his entire trajectory.
How did the Marine Corps shape his approach to leadership?
The Corps taught Victor that leadership isn’t a title — it’s what you do when no one’s watching. He learned to lead by example, make decisions under pressure, and trust his team to execute. That discipline carries directly into how he runs OUTURE and approaches every project.
What was the thinking behind founding OUTURE?
After the Marine Corps, Victor saw how businesses struggled with cross-cultural operations — not the logistics, but the human side. OUTURE started as a thesis: that consulting needs people who’ve actually lived across cultures, not just studied them. His background — Sweden, New York, the military — became the foundation.
What was the hardest part of the AI ethics case study?
Building the AI consultant persona was straightforward. The hard part was arguing against its recommendations — the model made a compelling utilitarian case for restricting patient access, and Victor had to dismantle it without dismissing the logic entirely. That tension between agreeing with an AI’s reasoning and challenging its conclusions became the core of the paper.
Latest-generation large language models with advanced reasoning capabilities
Autonomous system that plans, reasons, and decides which tools to use before responding
Retrieval-augmented generation over structured professional history and documentation
Professional history, education records, case studies, project documentation, certifications
Structured tool use for querying, cross-referencing, and surfacing specific information on demand
Natural language interaction — no menus, no navigation, just questions and answers
Real-time transcription that converts spoken questions into text for KAI to process
AI-generated speech built for KAI — responses you can hear, not just read
Full conversational pipeline from spoken question to spoken answer with minimal latency
A portfolio is a document. It compresses years of decisions, growth, and context into bullet points and section headers. The best professional interactions aren't reading exercises — they're conversations. You ask a question, you get nuance. You follow up, you get depth.
KAI is an experiment in what professional identity looks like when it can speak for itself — in its own voice, drawing on Victor's reasoning, grounded in his work. Not replacing the portfolio. Extending it into something you can talk to, and that talks back.
KAI is actively being built — the agentic architecture, knowledge base, and tool layer are all in development. When it launches, it will live at kiani.ai.
Interested in the project or want to know more?