Fanout Query Analysis
When AI models like Gemini, GPT or Nova answer a question using web search, they don’t just run your query as-is. They generate their own internal search queries, or fanout queries. A single user prompt can trigger multiple fanout queries as the model breaks down the question, explores subtopics and verifies information. We captured 365,920 […]
Reverse Prompting: Reconstructing Prompts from AI-Generated Text
We fine-tuned Google’s Gemma 3 (270M) to reverse the typical LLM workflow: given an AI-generated response, the model reconstructs the most likely prompt that produced it. We generated 100,000 synthetic prompt-response pairs using Gemini 2.5 Flash, trained for a single epoch on a consumer GPU, and built a Streamlit app that sweeps 24 decoding configurations […]
Rufus – Under the Hood. What Drives Amazon’s AI Shopping Assistant?
What’s Publicly Known About the Pipeline, Backend, and Response Anatomy. Rufus is not “one model that magically answers.” Public Amazon/AWS descriptions point to a multi-component system: Speculative schema: Pipeline: request → answer Step A — Input + context assembly Public descriptions indicate customers can: Amazon also describes using conversational context and (more recently) account memory […]
Search Grounding is Transient
There is a fundamental misconception about how Google’s AI search and Gemini chatbot process retrieved web content. It is widely understood that these systems use Retrieval-Augmented Generation (RAG) to search the web, pull snippets from pages, and ground their answers in factual data. However, there is a pervasive assumption that once an AI pulls in […]
What extraction method is Google using to build grounding snippets?
I’ve been reverse-engineering Google’s Gemini grounding pipeline (AI Mode, Gemini Chat…etc) by examining the raw groundingSupports and groundingChunks returned by the API. Specifically, I’m interested in the snippet construction step, the part where, given a query and a retrieved web page, the system selects which sentences to include in the grounding context supplied to the […]
Implicit Queries in AI Search
Back in 2015 I wrote about Google’s reliance of user behaviours signals for ranking purposes. In that article I already covered their use of implicit signals, but now there’s an update! While investigating Google’s grounding pipeline (the system that feeds web content to Gemini before it generates an answer) I came across the same patent […]
Bias and Prejudice in AI Search
When Claude Met DEJAN I was helping a developer debug a machine learning pipeline. Forty million training samples, weighted loss functions, checkpoint management — technical work. At some point, they asked me to generate test queries for their keyphrase volume classifier. I needed examples across the search volume spectrum, from high-volume head terms down to […]
Google’s Trajectory: 2026 and Beyond
AI is shifting from tool to utility. Agentic AI Becomes the Default Interface 2026 prediction: Expect Google Search to become agentic by default. Not “here are 10 links” – more like “I booked the restaurant, here’s the confirmation.” Operator-style functionality baked into Search and Gemini app. Gemini 4 Likely Late 2026 The pattern is clear: […]
Google’s Ranking Signals
Popularity Popularity signals are derived from user interactions based on ingested user events. The more the users interact with a document, the stronger the boosts are. These data requirements check the overall readiness of your events to generate the popularity signals. This is regardless of the specific search app that you choose. Predicted CTR model […]
How big are Google’s grounding chunks?
Dataset Metric Value Total Snippets 18,005 Total Fragments 883,262 Avg Words / Chunk 15.5 Dataset Vocab 12,728,193 Statistical Summary Chunks Words Chars Words per Chunk count 18005 18005 18005 18005 mean 49 706 4500 15 std 50 855 4681 9 min 1 4 18 1 25% 20 300 1923 12 50% 29 418 2648 […]
Google’s AI Uses Schema?
Article updated thanks to a sharp observation from Lukasz Rogala who makes my claim less certain and putting us back in the “needs more evidence category”. There’s some evidence Google uses structured data to ground Gemini in its AI search. If true this is good news for AI SEO people and vindication for schema advocates […]
Google’s AI Uses Schema?
Article updated thanks to a sharp observation from Lukasz Rogala who makes my claim less certain and putting us back in the “needs more evidence category”. There’s some evidence Google uses structured data to ground Gemini in its AI search. If true this is good news for AI SEO people and vindication for schema advocates […]
Google’s AI Uses Schema?
Article updated thanks to a sharp observation from Lukasz Rogala who makes my claim less certain and putting us back in the “needs more evidence category”. There’s some evidence Google uses structured data to ground Gemini in its AI search. If true this is good news for AI SEO people and vindication for schema advocates […]
Dynamic Visual Layouts
Dynamic visual layout (DVL) is a class of generative user interface which acts as an ephemeral information substrate. For two decades, SEO has been about fitting information into layouts. The blog post template. The product page schema. The FAQ accordion. The listicle format. We optimized content for containers that existed before the content did. Google […]
Grounding Snippet Extraction Tool
You can rank #1 and still be invisible to AI search. That’s the uncomfortable truth of the AI Mode era. Google’s AI doesn’t just look at your page, it extracts specific sentences, evaluates them against the query, and decides whether your content deserves to ground its answer. The rest of your carefully crafted copy? Find […]
Grounding Snippet Extraction Tool
You can rank #1 and still be invisible to AI search. That’s the uncomfortable truth of the AI Mode era. Google’s AI doesn’t just look at your page, it extracts specific sentences, evaluates them against the query, and decides whether your content deserves to ground its answer. The rest of your carefully crafted copy? Find […]
How user prompts shape your content visibility in AI search.
A single article. Seven different queries. Radically different passages surfaced. This isn’t a bug. It’s the ranker doing exactly what it’s supposed to do—and it reveals something important about how content actually gets discovered in AI search. The Data We ran seven query variations against one health article about teas for ulcerative colitis. The article […]
Advanced Prompting Techniques for AI SEO
Most marketers treat AI like a magic box: prompt goes in, content comes out. But AI models are more like highly skilled interns—they need clear instructions, context, and examples to do their best work. The quality of your AI output is directly determined by the quality of your prompts. Master prompt engineering, and you can: […]
To block or not to block? Bot is the question.
Are you accidentally slamming the door on helpful AI visitors while trying to keep your website’s content safe from being scraped for training data? Many site owners block bots to protect their intellectual property, but in doing so, they might be turning away the “good” AI traffic—like search engines and assistants that drive real visitors […]
Gemini 3 hallucinates fan-out queries
TL;DR: Gemini 3 made up the fan-out queries used to answer a prompt. Today I was testing the updated API response from Gemini 3 (thanks Mike!) and found it to be as unreliable as its predecessors when it comes to hallucinations. Not only did it lie to me, but it also attempted to cover up […]