Observability with real insights and auto-fixes (cloudgrip.ai)
Ex150ish-Fruit-and-Chips (theheartattackdiet.substack.com)
Show HN: Seekeasy – MCP for Restaurant Recs Powered by Social Media Data
We're building search over social media data - starting with the restaurants vertical first but we plan to cover all lifestyle topics (travel, fashion, beauty, sports etc).
To date, we've ingested 1M+ instagram posts (images and videos), 7K+ food creator profiles, and 150K+ restaurants data across SF and NY (expanding nationwide very quickly).
We got a lot of interest in our dataset so we built an MCP for devs to explore our data where you can query for restaurant recommendations and see results based on relevancy + instagram videos supporting the result (e.g. "first date spots"). We're adding ~1K instagram posts daily and will start ingesting TikTok soon as well.
Additionally we're working on time series analyses to better understand trends such as newly opened restaurants, which restaurants are popular now, creators that are "up and coming"... We are thinking to expose these as APIs or more MCPs for other client applications.
Would love to see if this sort of data is useful to others and hear what other use-cases you might want out of this.
Thanks!
Other links:
* About Seekeasy: https://about.seekeasy.ai/
* Our consumer app if you'd like to give it a spin (all feedback and roasting welcome): https://app.seekeasy.ai/download
* Product Hunt launch: https://www.producthunt.com/products/seekeasy-mcp
Couple thoughts: • Would kill for a trends API. • Would also love a way to surface rising creators based on post frequency + engagement velocity — could plug right into a rec engine or UGC curation pipeline. • If you expose the ingestion layer down the line, sign me up. Been dreaming of a “ChefGPT” that spits out recs based on how a dish looks on IG.
Anyway, very cool. Subscribed, cloned, slightly obsessed. Let’s see what I can hack together this weekend.