Tebak Kata

Contexto-inspired semantic word-guessing game built with Next.js and vector embedding similarity matching

Tebak Kata is an interactive semantic word-guessing game (inspired by Contexto) built with Next.js and high-dimensional vector embedding similarity matching.

Private Project Repository

Features & Engineering

  • Semantic Embedding Matching: Computes cosine similarity between player guess embeddings and the secret target word using dense semantic representations.
  • Dynamic Proximity Ranking: Instant rank sorting and visual proximity feedback based on semantic distance in embedding space.
  • Modern Next.js Architecture: Fast, reactive client-side rendering with serverless API routes for embedding lookups and score persistence.