Applied AI Researcher
Oshen Davidson
I use data science and machine learning to understand LLMs' behavior and the impact it has on society.
My Thoughts on AI
LLMs are evolving at a pace that is reshaping how people find information, make decisions, and do their work. I find that both exciting and sobering. The acceleration means new capabilities surface faster than most people can evaluate them, and the systems generating those capabilities are already influencing what people believe, what they act on, and how much they can trust what they're reading. That's why I study how these systems retrieve, select, and generate information. Part of that work is practical: helping businesses and brands understand how they show up in AI-generated answers.
But the larger reason is that understanding how these systems actually function is the foundation for understanding their risks, their security implications, and the ethical questions they raise about accuracy, transparency, and accountability. These systems are already mediating how people navigate their health decisions, their finances, their children's education, their understanding of the world around them, whether they realize it or not. I want to be part of making sure that mediation is studied, understood, and held to a standard that protects people while still giving them access to what this technology makes possible.
Surviving the Citation Gate: What ChatGPT Cites When Buyers Are Asking
What separates pages that ChatGPT cites from pages it retrieves and discards, conditioned on commercial intent. 217,508 retrieved pages across 7,500 commercial prompts, segmented by buying-journey stage and analyzed under four cross-checked analytical passes.
Inside ChatGPT's Search-to-Citation Funnel: Where Pages Lose Visibility
How ChatGPT moves from search to citation. 548,534 pages retrieved across 15,000 prompts analyzed to separate the retrieval decision from the citation decision, with rank-controlled six-model decomposition and DA × rank tier cross-stratification.
Community Platforms and UGC as Citation Sources in AI Search
How user-generated content and community platforms shape brand visibility across ChatGPT, Perplexity, Gemini, and Google AI Mode. 5.5M answers and 61K queries analyzed across 821,000 cited domains.
Selected Work
Content Quality Data Pipeline
15,000-query empirical research pipeline measuring which content quality signals predict ChatGPT citation. Six idempotent pipeline steps collect ChatGPT responses through the real consumer interface, join against Google SERP data for original and fan-out queries, extract 24 content quality signals per page, and run multi-method statistical analysis.
AirOps Research Agent System
19-agent, 13-skill multi-agent orchestration system running end-to-end research report production at AirOps. Plugin-style YAML-defined agents with per-agent skill loadouts, three lifecycle maps (full report, micro-report, insight drip), per-agent session memory, and cross-tool MCP integration across Notion, Slack, Asana, AirOps, ClickHouse, and Google Workspace.
Fan-Out Coverage Study
16,877-query observational research pipeline measuring whether ChatGPT cites pages that cover more of its internal fan-out sub-query space, against a deterministic seen-but-not-cited control group. Queue-orchestrated multi-worker pipeline with novel headings-as-index coverage scoring and on-demand section-text embedding.
As Featured In
- Zyppy
- Search Engine Land
- MSN
- eMarketer
- MSN Deutschland
- Forrester