Predict the next pint.

The fourth pillar of x-brain — cannabis (BudBrain), wine (WineBrain), food (FoodBrain), and now beer. Same engine, beer-domain features.

What BeerBrain does

• Snap a can, get a prediction. Single-can vision flow extracts brewery, beer name, style, ABV, IBU, and infers perceptual feature scores (bitterness, body, hop intensity, roast, funk).
• Snap a taplist, get N predictions. The chalkboard at your local taproom becomes 12 ratings ranked by how much you'll like each beer.
• Style-aware similarity. Cosine over abv/ibu/body plus indicator features (style_neipa, hop_citra, yeast_brett) keeps a stout from being recommended after you only loved IPAs.
• Brewery context matters. Same beer style, different brewery — the predictor learns your preference per producer.
• Cross-brain pairings. Friends who loved this NEIPA ate this BBQ; friends who rated Pilsner Urquell at the bier garten loved this schnitzel. Powered by the x-brain knowledge graph.

How the prediction works

Same hybrid engine that powers BudBrain, WineBrain, and FoodBrain: a 15-question style-discriminator onboarding ladder, k-means cohort assignment, content-based fallback over the beer feature schema, and a per-user XGBoost model that takes over after ~20 ratings.

Where the catalog comes from

The Pennsylvania flagship lineup is seeded — Tröegs, Yards, Victory, Iron Hill, Weyerbacher, Sly Fox, Tired Hands, Forest & Main, and a dozen others. Open Brewery DB extends the brewery directory nationally. Per-brewery JSON-LD scrapers fill in detail rows for the breweries that publish structured product metadata. Anything missing? Snap it — the can-vision flow seeds new items as you encounter them.