Live data from Hacker News

Viewing profile — tylersuard

tylersuard

HN member
Joined
Mon, Jul 16, 2018, 11:13 PM UTC
HN karma
17
Public activity
19 items

About tylersuard

No profile information was provided.

Recent public activity

  1. comment
    Comment #48901967

    I made a neural network with physical knobs on it so I can get the feel for training a network.

  2. story
  3. comment
    Comment #48704464

    Learn how AI works by training your own neural network by hand.

  4. story
  5. comment
    Comment #47084136

    Hi HN, I built a draft protocol called AIP (Agentic Internet Protocol). The premise: instead of trying to make agents use human websites, build agent-only websites from the ground …

  6. story
  7. comment
    Comment #44625455

    Thank you for the correction. I just assumed that the problems with the software were down to laziness on behalf of the devs, but from what you are saying, the devs are overworked …

  8. comment
    Comment #44622980

    Blog-rant about how much I hate Azure.

  9. story
  10. story
    Show HN: MCP-123, a 2-line MCP server/client (Windows-friendly)

    Got tired of every MCP example being overly verbose, or needing Docker or Mac-only scripts, so I threw together MCP-123. Point it at a tools.py, run `server.run_server(...)`, and t…

  11. comment
    Comment #43722509

    For other students reading this, here is how you can work as a software developer with no experience: start your own company (just make a website for it) and work as a freelancer o…

  12. comment
    Comment #43439502

    Ah, sorry, I forgot to mention it. The title will be "Enterprise RAG: Scaling Retrieval Augmented Generation"

  13. comment
    Comment #43439497

    Great, thank you! The main program is hosted on Azure Web Apps, the search is Azure AI Search, we use AutoGen for the agents, and we use OpenAI for the generation. Azure has a lot …

  14. comment
    Comment #43439471

    Ok so the company has like 20 databases, plus over 100,000 pages of PDF catalogs. We tried using agents to query the company's SQL databases, but that took 30 seconds each call, an…

  15. comment
    Comment #43439443

    First of all, great question. Second, we use a search service, and vectors are treated as supplementary to the text search, so chunking doesn't matter as much. We will usually take…

  16. comment
    Comment #43439428

    We have tests in place to make sure every function, every search is working, otherwise things won't deploy. We do have a dashboard with usage etc., but I have not spent much time l…

  17. comment
    Comment #43439397

    EXCELLENT question. We tried about 5 different ways of retrieving data, and we found that what works best for us is a search-as-a-service. We use Azure AI Search, but there are lot…

  18. comment
    Comment #43439368

    How to start small in a very specific area: 1. Write down a list of 10 to 20 questions that you want your RAG to be able to answer. Then write down the correct answers next to each…

  19. story
    Ask HN: Lessons from Building a Fortune 500 RAG Chatbot (50M Records in 10–30s)

    I’ve spent the past year and a half constructing a Retrieval Augmented Generation (RAG) chatbot for a Fortune 500 manufacturing company, integrating over 50 million records across …