← Home

Work

I don't just study AI search, I build the products and run the experiments that test what actually gets cited.

Projects

★ Flagship project Flozi ↗

Platform designed to analyze how websites perform in AI search systems and provide guidance for improving retrievability and citations.

AI search visibility retrievability analysis AEO diagnostics
Experiment

Embedding Similarity Analyzer

System that analyzes semantic similarity across website pages using vector embeddings and cosine similarity to detect redundancy and topic drift.

vector embeddings cosine similarity semantic clustering
Internal System

AI SEO Execution System

AI-assisted workflow system for executing SEO tasks faster using modern AI tools combined with traditional SEO platforms.

AI automation SEO execution content optimization

Research

Pilot Study · GPT-4o · April 2026

Beyond Rankings: Measuring Vendor Visibility in AI-Driven Discovery ↗

Introduces ISIC — a four-dimensional probabilistic visibility model (Inclusion, Stability, Influence, Coverage) — as the measurement framework for AI-driven discovery. Empirically validates that LLM and Google search channels direct buyers toward fundamentally different vendors, with an average Jaccard overlap of 0.03 across 27 B2B CRM queries.

AEO measurement probabilistic visibility LLM citation bias B2B discovery

Experiments

AI Citation Testing

Testing how different content structures influence whether AI systems cite a page in generated answers.

Retrievability vs Authority

Exploring why pages retrieved by search engines are not always cited by AI systems.

Semantic Content Mapping

Using embeddings to visualize relationships between pages and detect content gaps.