Seltz
The fastest, high quality, search API for AI agents
Details
- External ID
- 47837377
- Source
- HN
- Company
- —
- Product
- Seltz
- Website domain
- seltz.ai
- Launched
- April 20, 2026
- Cohort
- —
- Upvotes
- 5
- Upvotes percentile
- 0.11182519280205655
- Tags
- —
- Fetched at
- Sept. 7, 2026, 9:26 p.m.
- Updated at
- Sept. 7, 2026, 9:26 p.m.
Description
Hi HN, Antonio here. Founder of Seltz.Seltz is a web search API built for AI agents. We wrote the crawler, the index, and the retrieval models ourselves, in Rust, by a team that's spent years building web search at scale. In our tests, queries come back in under 200ms.Efficiency was the first design principle. Search sits on the critical path: agents can't generate their first tokens or kick off the next tool call until results come back. When you run tens or hundreds of queries in parallel, every millisecond of tail latency compounds.Most search APIs for agents are wrappers around Google or Bing. If your agent already has a Google tool, a second call to a Google-wrapped API returns the same ten documents. We run our own independent index, so you get different results and different rankings.Coverage starts with US news. More verticals are coming.There's a free tier with $100 in credits at the link. I'd love feedback from anyone building agents: how does it compare to what you're using, and where does it fall over?
Enrichment
- Theme
- browser automation and scraping for AI
- Vertical
- Horizontal
- Function
- Search & retrieval
- Audience
- Developer
- AI stance
- AI feature
- Project type
- Commercial product
- Normalized one-liner
- search api for ai agents
- Manually corrected
- False
Could you build this?
No Building a web-scale search engine with custom crawlers, inverted index, sub-200ms retrieval, and web data pipelines requires huge infrastructure capital and advanced search systems engineering.
What it would actually take: Building Seltz requires a massive distributed crawler infrastructure capable of handling politeness, JavaScript rendering, and petabyte-scale storage, coupled with an inverted index and vector search engine written in Rust or C++. The hardest parts are maintaining fresh web-scale crawl indexes, deduplication, spam filtering, and low-latency distributed query retrieval under 200ms. It requires millions in bandwidth/compute and a veteran search/information retrieval engineering team.
Discussion
1 comment analyzed.
Competitors mentioned: 10 other search APIs (unnamed), PISA Engine
Concerns raised: Latency in agent loops limits follow-up queries, Search accuracy benchmarking needs rigorous methodology
Competitors
Other products that read as similar to this one — 125 launches clear the similarity bar, closest 8 shown.
Attention rank: #111 of 126 (itself plus its competitors, highest first — normalized so YC and Product Hunt are compared fairly).
Launched 164 days after the earliest competitor.
- Keenable · hn · 2026-08-25 · 12 upvotes · similarity 0.51
- The new Firecrawl /search · ph · 2026-07-24 · 259 upvotes · similarity 0.51
- Telem · hn · 2026-08-27 · 8 upvotes · similarity 0.46
- Crustdata (YC F24) · hn · 2026-02-25 · 10 upvotes · similarity 0.45
- Crustdata - Web Search API for AI Agents · yc · 2026-02-05 · 79 upvotes · similarity 0.43
- Moltis · hn · 2026-02-12 · 131 upvotes · similarity 0.43
- Web Search Agents by Nimble · ph · 2026-09-14 · 306 upvotes · similarity 0.42
- Semble · hn · 2026-04-26 · 7 upvotes · similarity 0.42
Other launches for this product
- No other launches for this product.