E-commerce

Rank vehicle listings from multiple sources

Use case: Pull live used EV listings from eBay Motors and Carvana, normalize them into a single dataset, score each listing on five weighted factors, and explore the results in a filterable Streamlit dashboard with direct links to live listings.

Quick Start

Inputs

  1. EV models Vehicle makes and models to search on eBay Motors and Carvana, for example Tesla, Nissan Leaf, Chevy Bolt, and Kia EV6.
  2. Filters Make and model, year range, max price, max mileage, and minimum deal score, applied in the dashboard sidebar.

Outputs

  • Used vehicle listings from eBay Motors and Carvana, scored and ranked
  • Deal score (0–1) per listing, based on 5 weighted factors
  • Price vs. mileage scatter chart by deal score
  • Deal score distribution by model
  • Filterable listings table with links to live listings
  • CSV export

818 scored listings from eBay Motors and Carvana, collected April 30, 2026. No API key is needed to explore the dashboard.

View example on GitHub

How it works

A 6-phase pipeline.

  1. Search The eBay search agent pulls listing IDs across 10+ vehicle model queries, returning titles, prices, and listing URLs for each match.
  2. Extract All eBay listing pages are fetched in parallel to pull structured item details: make, model, year, mileage, condition, and title status.
  3. Build agent A custom Carvana agent is built from scratch using Nimble Agent Builder. No manual selectors are needed; the builder handles pagination and render timing automatically.
  4. Collect The Carvana agent paginates through 30+ pages of live EV inventory and returns structured listing data per page.
  5. Normalize 818 listings from both sources are merged into a unified schema: source, make, model, year, price, mileage, condition, title status, and listing URL.
  6. Score A 5-factor deal score is computed per listing: price (30%), mileage (25%), year (20%), condition (15%), and title status (10%), scored relative to the full dataset.

Stack

Nimble primitives plus the full runtime stack.
Nimble APIs
What it does
  1. Nimble Agent Builder Used to create the Carvana agent from scratch; schema defined visually, no CSS selectors written.
  2. ebay_search_results_community_2026_03_23 Pulls eBay Motors search results and returns item IDs for individual listings across any EV model query.
  3. extract_batch Batch-fetches individual eBay listing pages in parallel, extracting Item Specifics including make, model, year, mileage, condition, and title status.
  4. carvana_electric_cars_serp_2026_04_30_auw1p8rf Custom agent built with Nimble Agent Builder that paginates Carvana's EV inventory and returns structured listing data per page.
3rd Party Tools
Role
  1. streamlit Dashboard with sidebar filters, scatter/box/bar charts, and a ranked listings table.
  2. plotly Price vs. mileage scatter chart, deal score distribution by model, and listings by year bar chart.
  3. python 3.9+ Scoring engine, data normalization across two sources, and batch extract orchestration.
Reach out if you have any questions.
Talk to an Expert