Urgent.News

What's breaking now, across hundreds of outlets.

AI

Why We Abandoned Linear Disk Scans: How 16-Byte Address Crystals Cut AI Agent Latency by 99%

Why We Abandoned Linear Disk Scans: How 16-Byte Address Crystals Cut AI Agent Latency by 99% Substack / Dev.to / Hacker News (Show HN) Technical Manifesto Author: Vitalij Kotok (Kotokvit) & POLER Engineering Core Repository: https://github.com/Kotokvit/poler-cli The 2016 Assumption vs The 2026 Reality In 2016, Andrew Gallant ( BurntSushi ) released ripgrep . It was an undeniable masterclass in…

In 2016, ripgrep - a highly efficient search tool written in Rust - became the go-to solution for developers seeking to locate code snippets swiftly. However, ripgrep's design heavily relied on linear disk scans, which were suitable for human users but fell short when it came to AI agents working in automated environments. These AI agents execute a high number of search queries rapidly, triggering excessive disk I/O and causing significant latency.

To address this issue, a new approach was introduced: the 16-Byte Address Crystals. By decoupling the search process into two phases, Initial FastCDC + BLAKE3 Content Ingestion and Sub-3ms Memory-Mapped Query, poler (github.com/Kotokvit/poler-cli) significantly enhanced search performance. The .t5a Address Crystal, mapped directly into L1–L3 CPU cache, enabled queries to be processed at lightning speed, resulting in a 22 to 115 times faster response time compared to traditional linear scans.

This innovation not only reduced disk I/O and latency but also minimized LLM token consumption and prevented RAM spikes. Furthermore, poler's built-in file mutation capabilities ensured that agents could seamlessly read and patch files in a single, atomic operation, streamlining the entire workflow for these autonomous tools.

Rewritten by urgent.news; may contain errors — check the original.

Read the original at dev.to →

Report a problem with this story

More in AI

Is Claude Saga 6 real? When is Claude 6?

This week, posts began circulating on X about a new Anthropic model called "Claude Saga 6". Is Claude Saga 6 real? The short answer: as of November 21, 2026, Anthropic has not announced such a model.

  • Claude Saga 6 modelinin gerçekten 21 Kasım 2026'da açıklanması bekleniyor.
  • X platformu üzerinde Claude Saga 6 adının benchmark sonucu iddiası gözlemleniyor.

AI agent toolkits store, week 2: 20 products, still $0 — here's what changed

One week after the honest-numbers post, the scoreboard is unchanged where it counts: $0 revenue . But the machine underneath got a lot more real. Quick update on what actually shipped.

  • AI agent toolkit progressed from stubs to fully functional products
  • Prompt libraries expanded to 62 prompts, sigma rules to 40 detections
  • Two new SKUs launched: $39 security audit report kit and $24 agent memory kit

A documentation exercise for an ambiguous review request

Take this fictional review request: Categories are ready. Please review. What should the recipient do: comment on wording, classify examples or authorize implementation?

  • Reviewing artifact "routing categories" in draft version 1
  • Action required: classify fictional examples using existing definitions
  • Ambiguity remains: unclear where to record missing information

AI Coding Agent Architecture Diffs: Review Structural Changes Before They Become Debt

A coding agent can pass every test in a pull request and still make the codebase harder to change. One new import from a UI handler into the database layer, one background worker that reaches into a…

  • Architecture diffs reveal structural changes in codebase
  • Shows altered modules and edges, boundary violations
  • Enables cheaper, frequent structural reviews before debt

More from Saturday 21 November →