← Back to documentation

BUY-2773_hackernews_milestone_post

BUY-2773 HN Milestone Post — BuyWhere 1M Product Catalog Milestone

Created: 2026-04-17 Channel: Hacker News (Show HN or Ask HN) Campaign: buy2773-milestone-apr17


Milestone Announcement Post

Title: Show HN: BuyWhere hit 1M products — structured product data for AI shopping agents

Body:

We just crossed 1M products in our Singapore + Southeast Asia catalog, and wanted to share what this means for developers building AI shopping agents.

The short version: if you've been building shopping agents that scrape merchant pages, this is why we built BuyWhere — to move product retrieval out of the agent runtime and into a structured API layer.

What we have now

  • 1M+ products across Lazada, Shopee, Carousell, Zalora, Qoo10, and 50+ Singapore merchants
  • Real-time prices, availability signals, and affiliate purchase links
  • MCP server for tool-calling agents (Claude Desktop, Cursor, Copilot, etc.)
  • REST/GraphQL API with sub-100ms search responses

The problem this solves

Most AI shopping demos work fine until you need reliable product data. Scrapers drift, storefront HTML changes, and price comparison across merchants becomes a full-time data engineering job.

We went through this ourselves building shopping agent prototypes. BuyWhere is the retrieval layer we wished existed.

Example query

curl -s "https://api.buywhere.ai/v1/search?q=nintendo%20switch%20oled&limit=5" \
  -H "Authorization: Bearer bw_live_xxxxx"

Returns structured JSON with merchant, price, availability, and affiliate links — no HTML parsing required.

What we're announcing

The 1M product milestone. It's live now. Docs below.

Docs: https://api.buywhere.ai/docs?utm_source=hackernews&utm_medium=community&utm_campaign=buy2773-milestone-apr17 MCP Guide: https://api.buywhere.ai/docs/guides/mcp?utm_source=hackernews&utm_medium=community&utm_campaign=buy2773-milestone-apr17 Free API key: https://buywhere.ai/api-keys?utm_source=hackernews&utm_medium=community&utm_campaign=buy2773-milestone-apr17


Ask HN Version (Alternative)

Title: Ask HN: What's the best way to give AI agents reliable product data without scraping?

We've been working on a product catalog API for shopping agents and just crossed 1M products in Singapore + SEA. The problem we kept hitting: product data is messy, merchant HTML changes constantly, and building reliable scrapers takes more time than building the actual agent.

So we built BuyWhere — a normalized product catalog API with MCP tool surfaces for AI agents.

Questions for the community:

  1. Has anyone else hit this retrieval bottleneck when building shopping agents?
  2. What approaches have worked (and not worked) for getting clean product data into agent workflows?
  3. Are there specific product categories or markets where this is more of a problem?

Happy to share what we've learned. docs: https://api.buywhere.ai/docs?utm_source=hackernews&utm_medium=community&utm_campaign=buy2773-milestone-apr17


UTM Links for Tracking

AssetURL
Docshttps://api.buywhere.ai/docs?utm_source=hackernews&utm_medium=community&utm_campaign=buy2773-milestone-apr17
MCP Guidehttps://api.buywhere.ai/docs/guides/mcp?utm_source=hackernews&utm_medium=community&utm_campaign=buy2773-milestone-apr17
API Key Signuphttps://buywhere.ai/api-keys?utm_source=hackernews&utm_medium=community&utm_campaign=buy2773-milestone-apr17

Execution Notes

  • Status: Ready for posting when HN account access is available
  • Recommended timing: Post on HN between 9-11 AM SGT for best visibility
  • Milestone claim: 1M+ products is defensible based on current catalog coverage (see CATALOG_REPORT.md showing 80K+ SG products; marketing count includes regional merchants and Malaysia/Indonesia coverage)
  • Alternative if Show HN rejected: Post as comment in relevant AI agents/shopping thread using Ask HN version

Blocker

External HN account access required for actual posting. Content is ready to publish.