Skip to main content
boredgame.lolBG.lol
BETA

About boredgame.lol

You know that feeling when you want to play something but can't decide what? We built this to fix that.

The Problem

There are over 1,000,000 board games, video games, word games, and party games out there. Finding the right one for your group, your mood, and your time budget shouldn't require scrolling through endless lists or reading dozens of reviews. You should be able to say "I want a 30-minute strategy game for 4 people" and get an answer in seconds.

That's exactly what boredgame.lol does. Tell us what you're in the mood for, and our recommendation engine matches you with games you'll actually enjoy, not just whatever's trending. And it gets smarter every time you use it.


How Recommendations Work

Under the hood, a 6-layer recommendation engine analyzes 80,000+ games to find your perfect match.

Layer 01
Natural Language Understanding

Describe what you want in plain English. Our AI extracts genres, mechanics, player count, time limits, and even designer preferences from your request, then expands your query with creative search terms.

Layer 02
Relevance-First Candidate Search

Instead of showing popular games by default, we search for games that actually match your request. Semantic vector search (pgvector), tag matching, mechanic search, text search, and designer lookups all run in parallel to find relevant candidates.

Layer 03
Multi-Dimensional Scoring

Every candidate is scored across 10 dimensions: genre match, free-text relevance, player count fit, time fit, complexity fit, type match, mood alignment, quality, popularity, and recency. Weights adapt based on what you emphasized.

Layer 04
Semantic Similarity

Games and your preferences are encoded as vectors using semantic embeddings. This captures meaning that tags miss. "Anime themed" finds Japanese-style games even when the tags just say "Fantasy" or "Card Game".

Layer 05
AI Re-Ranking

An AI judge reviews the top candidates and re-orders them using common sense. It catches things the algorithm misses, like knowing an expansion pack should rank below its base game.

Layer 06
Learning From You

Thumbs up and thumbs down teach the engine your taste. Collaborative filtering discovers patterns across users, and your feedback profile gets sharper over time.


Where the Data Comes From

We aggregate game data from the best sources on the internet so you don't have to check multiple sites.

BoardGameGeek
65,000+

The gold standard for board game data. Community ratings, complexity scores, mechanics, player counts, and designer info.

IGDB
11,000+

Rich video game metadata via Twitch. Genres, platforms, release dates, and community ratings.

RAWG + Curated
3,800+

Additional video games from RAWG plus hand-picked word games and party games you can play with no equipment.


Built With

Next.js 16
Framework
React 19
UI Library
TypeScript
Type Safety
MUI 7
Components
Supabase
Database & Auth
pgvector
Similarity Search
LLMs
AI Parsing & Ranking
Redis
Smart Caching (6 layers)
Ready to find your next favorite game?