Nicheloom visits: 62,882 all-time · 8,415 in the last 7 days

The opportunity tracker for new startups.

Today: 1,441 launches · Niche of the day: digital marketing agency services · See what launched today →

Alcatraz

Pure-Go PII detection, 100x faster than MS Presidio

Get picks like this daily. Today's launch numbers, the top launches, AI and tech news, and a daily opportunity pick, straight to your inbox.

Want to reach these readers? Sponsor an issue.

Hi HN, I'm Andrios, founder of hoop.dev (YC W21), we build runtime controls for agents. We just released Alcatraz.We built it because our product is written in Go and we do real time Data Masking, but we were using MS Presidio (Python) for PII detection and it made connections slow.There was a good discussion here a few months ago when OpenAI released their Privacy Filter: (https://news.ycombinator.com/item?id=47870901). A lot of questions were built into our design. Structured identifiers (credit cards, SSNs, etc) are deterministic and you can verify them with checksum. Free text PII like names and addresses is where you actually need a model.Alcatraz started as our internal solution, but we wanted to make it available for everyone: in-process PII detection in Go. Currently 45 entity types across 12 countries (US and Brazil included). Benchmarks vs MS Presidio (lib to lib) about 13x faster on dense documents, 100x on smaller ones.Still in development, lots to improve. Feedback wanted: github.com/hoophq/alcatrazEdit 1: fix typo Python written as Phyton

HN · launched Aug 4, 2026 · 6 upvotes (top 69%) · 0 views

Niche: application and system security scanners · 271 launches, 22 in the last 7 days ▼ -49% · Explore this niche →

Category and tags

  • Security
  • Compliance & governance
  • Developer
  • Not AI
  • Commercial product

Source listing ↗

Could you build this?

Partial A basic regex-based PII scrubber is trivial, but matching or exceeding Microsoft Presidio's detection accuracy across diverse international entities purely in high-speed Go requires optimized NER and pattern recognition engines.

What it would actually take: The implementation requires Go-native high-performance tokenizers, Aho-Corasick or Hyperscan-backed regex matchers, checksum validators (Luhn, IBAN, SSN), and lightweight ONNX/NER models embedded via Go Cgo/pure-go bindings. The difficult part is matching production-grade recall/precision across unstructured text while keeping latency under 1 millisecond without heavy Python runtimes.

Discussion

No comments on this launch.

Competitors

Other products that read as similar to this one : 30 launches clear the similarity bar, closest 8 shown.

Attention rank: #22 of 31 (itself plus its competitors, highest first, normalized so YC and Product Hunt are compared fairly).

Launched 245 days after the earliest competitor.

The niche, labels, vibe code rating and competitor comparison on this page are analysis by Nicheloom. If you use them, please credit nicheloom.com. How you may use our data.