An AI-powered motorcycle service manual search tool. Parses chunked service manual data and expands part-name aliases to instantly surface torque specs, procedures, and references from motorcycle pdf service manuals.

Off-pipeline, low-confidence, and ambiguous extractions are routed into a review queue. Reviewers use it for human-in-the-loop validation, fixing bad parses, performing manual data cleanup, and adjusting synonym maps. Items are prioritized by a composite score that blends retrieval confidence and keyword-matching rank, so the highest-risk rows surface first.

Project Type: tech / PDF-to-Vector Search System

Role: Full-Stack Developer

Tech Stack (AI): llama3.2:3b, pdfplumber, all-MiniLM-L6-v2, Qdrant (Vector DB), Python, Custom Keyword Scoring Engine, Custom Synonym Expansion

Tech Stack (Web): Next.js, TypeScript, React, TailwindCSS, Vercel

Status: Deployed, 2026

Live Demo: moto-ref

motoref review queue

Off-pipeline, low-confidence, and ambiguous extractions are routed into a review queue