Scratch, Dent and Data: Engineering A Used Appliance Information Advantage
summary
Moves from manual calls to a data pipeline that scrapes 70+ used-appliance inventories across SC NC and GA, uses Google Maps/Places to build nodes and a 21x21 matrix of drive times via Distance Matrix, applies a greedy modified nearest‑neighbor heuristic to pick routes and closes (result: visits eight stores in a two‑hour window), and automates per‑site scrapers with LLMs (Find and Claude) to surface listings. Outcome: finds a 2008 Whirlpool direct‑drive set, gets delivered, and works.
AI-written summary · about
chapters (7)
- 0:00Introduction and laundry struggle
- 4:38Traveling salesman problem
- 10:38Data collection with GSP stack
- 14:52Heuristic solution code
- 17:59LLMs for web scraping
- 20:47Funny appliance store finds
- 24:58Leetcode philosophy in life
related talks
24mTransforming Data at a 110 Year Old Manufacturing Company
22mData Analysis in the Age of Data Democratization
31mBeginning your own data engineering projects
01h 03mDyalog APL: Our (Not So) Secret Ingredient
54mUsing Premium Data for Analysts
08mValue in Data Mashups
47mData Warehousing Migrations: Lessons from Home Depot
42mScraping Your Way to a Dataset