Hacker News

New stories

Live mirror
30 storiesupdated just nowView source snapshot
  1. Stages of (Developer) Grief (costafotiadis.com)
    —discuss
  2. As a freelancer, what happens when the door won't open any more? (lajosmoczar.substack.com)
    —discuss
  3. Nginx-Based RTMP Media Streaming Server (github.com/arut)
    —discuss
  4. Intel patent embedding MicroLEDs directly into CPU as an 'aesthetic component' (tomshardware.com)
    —discuss
  5. Consciousness as a Memory System (ovid.com)
    1comments
  6. Quantum Computer Completes in 19 Secs What Could Take a Supercomputer a Century (sciencealert.com)
    —discuss
  7. Martin Fowler: AI needs a conscience more than consciousness (martinfowler.com)
    —discuss
  8. Hierarchical Voxel Irradiance Caching [video] (youtube.com)
    —discuss
  9. Oxilite: An Oxigraph-compatible RDF database and SPARQL engine that uses SQLite (github.com/volland)
    1comments
  10. Show HN: Mecatl – Separate what an agent can do from where it runs (github.com/stacklok)
    —discuss
  11. Delhi Cut Electricity Loss from 50 to 5 Percent (ieee.org)
    —discuss
  12. Using an LLM to generate podcasts based on my journals, finances, & sleep data (jaan.io)
    —discuss
  13. Can AI get out of the way? (trysojourn.app)
    —discuss
  14. Russia Lights Rockets with an Oversized Wooden Match (2016) (popularmechanics.com)
    —discuss
  15. Binomial Coefficient Coincidences (utexas.edu)
    —discuss
  16. Show HN: Search2o – Search that finds and runs an AI agent for each request (pypi.org)
    —discuss
  17. Condé, Hearst among 300 media execs to push federal bad bots bill on AI scraping (digiday.com)
    —discuss
  18. Without the Hot Air (withouthotair.com)
    —discuss
  19. Zoom In: An Introduction to Circuits (2020) (distill.pub)
    —discuss
  20. Ask agent to find relevant investors (fundraisly.com)
    —discuss
  21. 1 in 8 cancer cases worldwide are caused by infections, study finds (cbc.ca)
    —discuss
  22. Self-host your own AI text detector on CPU to filter out slop (github.com/pablocaeg)
    —discuss
  23. I can spam anyone about the OpenAI DevDay
    —discuss
  24. Show HN: 5 mins daily programming quiz to prevent skill rot from AI coding (fata.app)
    —discuss
  25. Show HN: NotchMind – a native Mac app that puts the MacBook notch to work (notchmind.com)
    —discuss
  26. Building a Skill Interview with AI (hugoj0s3.dev)
    —discuss
  27. Dataviz: Most mentioned companies in the NYT since 01/01/2000 (habibicode.org)
    —discuss
  28. Lego Sorting via F# WebAssembly (johansivertsen.com)
    —discuss
  29. Self-Play Pretraining with Zero Data (arxiv.org)
    —discuss
  30. Show HN: CodeDiff – Fast (<100ms), robust (99.95%) syntax-aware code diff (github.com/ivankovic)
    —discuss

Show HN: Search2o – Search that finds and runs an AI agent for each request

1 pointsby 12m agopypi.org
0 comments
I have been working on search systems for a while, and I always wondered: What if search, instead of matching documents, matched workflows and ran those workflows?

I called it “Search that executes” — as opposed to search that retrieves information.

When AI agents showed up a few years ago, I realized that those workflows could be agents — capable of understanding a user’s query and taking branches dynamically. That’s when I got excited enough about the idea to implement it.

To make this work, it needs more than just search: a structured framework for defining agents (so the system “understands”), a controlled runtime for executing them with guardrails, and an end-to-end system with a UI to request and receive user approval before taking critical actions.

It made sense that all this plumbing should be handled behind the scenes, letting programmers concentrate on the agent’s business logic. Many design decisions followed directly from that.

Agents are created in a JSON structure. There are 23 “commands” that define the units of work an agent performs. Python expressions can be used in strings within curly braces.

Because the whole thing has a structure/schema, AI generates valid agents from English descriptions: “Create a RAG agent to call the company purchase order system”, “Create a tool-calling agent to …”. You can check this out in the live demo on the website.

The developer creates an agent, describes it in plain English, and publishes it in the system. Users just submit a request. Search matches the request to an agent, and the system runs the agent. The demo also shows searching for and executing agents.

That’s the central idea. The website has more information on other aspects. Despite the enterprise-y look and feel of the website, the system was really created by just me over the last couple of years. I would love to answer any questions and hear your comments and feedback.

A quiet thread, for now.Start the conversation on HN ↗