SEO Noobs

# Niche Scout AI — Evidence-Driven Niche Research and Validation I'm an independent SEO builder from Multan, Pakistan. I build niche websites around calculators, converters, checkers, and lookup tools, targeting underserved languages and countries. The hardest part of my work was never building websites. It was finding opportunities worth building. Traditional niche research required hours of manually checking keywords, search results, domain authority, search volume, competition, and market risks across multiple tools. **Niche Scout AI is the research agent I built to solve that problem.** Users select a country, language, and interest. The agent expands these inputs into approximately 250 localized utility search queries, analyzes search results, identifies weak competition signals, and evaluates surviving opportunities using available evidence. Its research pipeline combines keyword volume, domain authority, SERP features, seasonality, long-tail keyword clusters, and competitor content gaps. Muse Spark adds two reasoning capabilities: - **Judge:** Evaluates collected evidence and produces a plain-language verdict explaining whether an opportunity appears winnable, why, and what its main risks are. - **Plan:** Generates an actionable build brief containing positioning, recommended pages, content requirements, competitor weaknesses, and initial implementation steps. The platform uses transparent qualification criteria rather than blindly chasing high search volumes. Its demand gate considers keyword families above 50,000 monthly searches, while its competition gate looks for at least three weak domains in the top ten and no unassailable competitor in the top three. Other factors, including zero-click risk, seasonality, and government competition, remain visible as risk signals. In our validated Mexico calculators run, only one of twenty opportunities passed both gates: **calculadora de finiquito**, with 90,500 monthly searches and seven weak domains in the top ten. A currency converter with 301,000 searches was rejected because established competitors dominate its search results. The project also addresses practical AI deployment constraints. Since Meta's hosted Model API was unavailable in Pakistan for this project, Muse Spark Open runs through Baseten partner inference. Staged token budgets, retries, evidence-only prompts, and per-niche caching help manage reasoning-model limitations. When live model access is unavailable, the product clearly identifies saved results instead of pretending a new AI assessment occurred. Niche Scout AI is built from my real experience as an indie SEO builder. Its goal is to turn fragmented SEO research into structured, evidence-driven decisions that solo entrepreneurs can use to build better online businesses.