The reasoning layer between spreadsheets and strategy
My AI Analyst is the system I’m building at Sway Interactive — a structured intelligence layer that sits between raw data and real decisions. It maps datasets into semantic structures, detects statistical signal, and generates narrative explanations with traceable logic.
The goal is simple: move people from “what are these metrics?” to “what does this actually mean for the decision in front of me?” without requiring a full analytics team or a stack of dashboards.
How it works
The system starts by mapping tables into entities, metrics, and relationships — so it can reason about players, teams, portfolios, or cohorts, not just columns and rows.
It runs real statistical checks to find trends, outliers, and meaningful group differences, distinguishing between weak noise and genuinely notable patterns worth paying attention to.
Instead of dropping you into a wall of charts, My AI Analyst assembles clear, defensible narratives — the kind of explanation you’d send in a memo, not a screenshot you hope people interpret correctly.
LLMs are treated as components inside the system — used for interpretation, hypothesis generation, and language — while the underlying analysis remains explicit, inspectable, and grounded in data.
Where it’s headed
Early focus areas include sports analytics and portfolio management — domains where people constantly reason under uncertainty, weigh trade-offs, and need to explain their decisions. But the architecture is general on purpose: the same system should eventually power investment memos, strategic reviews, and other decision artifacts.
Long term, My AI Analyst is meant to feel less like a tool and more like a thinking partner: rigorous about data, honest about uncertainty, and fluent in narrative.
Contact