Hacker News

New stories

Live mirror
30 storiesupdated just nowView source snapshot
  1. Building the New GitHub Copilot Inline Suggestions Model: Part One(visualstudio.com ↗)
    discuss
  2. Senior Product Manager(talents2germany.de ↗)
    1comments
  3. Nullsec – server-blind encrypted paste(pntr.dev ↗)
    discuss
  4. EU Commission proposes under-13s social media ban(reuters.com ↗)
    discuss
  5. Equity Price Shock Scenario Would Entail US Recession(fitchratings.com ↗)
    discuss
  6. What Shall We Do with Broken Glasses?(komando1.substack.com ↗)
    1comments
  7. Jev Powered Knowledge and Agent Routing(amithkk.dev ↗)
    discuss
  8. Programming lanugage that uses procedures(fodev.net ↗)
    discuss
  9. The Dawn of Gestural Computing(asymco.com ↗)
    discuss
  10. Running a personal website is not fun anymore(winnielim.org ↗)
    2comments
  11. Unicode 18.0.0 has been released(reddit.com ↗)
    1comments
  12. Nolan Bushnell (Atari, Co-Founder) Career Interview (2006) [video](youtube.com ↗)
    discuss
  13. The DeepMind Institute(deepmind.com ↗)
    discuss
  14. Deriving neural scaling laws from the statistics of natural language(arxiv.org ↗)
    discuss
  15. The LLMs Yearn for the Spines(buttondown.com/hillelwayne ↗)
    discuss
  16. Lattice Prompt brings AI agents into the full FPGA design flow via MCP(businesswire.com ↗)
    discuss
  17. Semantics for 2D Rasterization(arxiv.org ↗)
    discuss
  18. Device Bound Session Credentials in Asp.net Core(blog.maartenballiauw.be ↗)
    discuss
  19. Embeddable Interactive Avatars(synthesia.io ↗)
    discuss
  20. The paradox of AI automation at the workplace
    discuss
  21. I didn't sign the Fields medallists' letter(gowers.wordpress.com ↗)
    5comments
  22. Artificial Analysis Capability Indices v1.1(artificialanalysis.ai ↗)
    discuss
  23. Show HN: AskMe – answers repetitive support questions(ask-me.studio ↗)
    discuss
  24. Automattic execs signed reciprocal severance deals in Mullenweg absence(techcrunch.com ↗)
    discuss
  25. Ask HN: How do you use AWS AssumeRole?
    discuss
  26. Reckless Expansion Plans Blow Past European Airports Carbon Limits(cleantechnica.com ↗)
    discuss
  27. What Is a Docker Container?(freipaul.com ↗)
    discuss
  28. AI Safety Is Mostly a Sex Cult(skywriter.blue ↗)
    10comments
  29. Stress at work linked to erectile dysfunction, even in younger men(scimex.org ↗)
    1comments
  30. Show HN: Airy TTS – 2K characters/second, $2 per 1M characters(airy.so ↗)
    discuss

Axiom Knowledge graph: A deterministic search engine

3 pointsby 1h ago
2 comments
Every time you search on internet you might feel in any corner of your mind as does this a correct information because now in 2026 every search becomes AI search and as we know AI can show errors and shows misleading information confidently which is wrong. We want a proper search engine which can show real facts and true facts with under millisecond of search. So, presenting Axiom a search engine which can solve these all problems. Axiom is not a simple search engine but a hybrid BM25 and a sematic vector search engine. AXIOM addresses these systemic pathologies by re locating computational control to a sovereign local exe cution node. AXIOM operates independently of external cloud services, processing web documents through a high-performance pipeline and indexing them into a locally stored vector space. AXIOM is explicitly architected for environments requiring high throughput, structural permanence, and absolute data privacy: • Offline-First Academic Research: Enables continuous, local ingestion of scientific repositories (e.g., ArXiv, PubMed, IEEE Xplore, CERN, Springer) for local query resolution without web access. • Deterministic Agentic RAG: Serves as a zero hallucination ground-truth context provider for autonomous AI agents requiring verified factual lookups. • Self-Hosted Knowledge Infrastructure: Provides high-density indexing of sensitive internal documentation alongside public domain datasets within a strictly air-gapped security perimeter. AXIOM presents a production-grade alternative to centralized search platforms. By unifying high-concurrency Rust spidering, local vector indexing in Qdrant, and strict cross encoder thresholding, AXIOM delivers high throughput information retrieval while ensuring absolute data sovereignty and factual accuracy. Axiom had won from Bing copilot search in information from Wikipedia about maths formulas and concepts and here Bing copilot search told this completely wrong and incorrect whereas on Axiom it shown correct and actual information.