Scott and Mark learn AI | BRK255

แชร์
ฝัง
  • เผยแพร่เมื่อ 21 ก.ย. 2024
  • Join Scott Hanselman and Mark Russinovich as they use AI tools to make smarter apps on Windows. You will learn how to use AI in your app, consume models both local and in the cloud, and how to use LLMs to make your apps even more awesome for users. They may even clean up Scott's messy desktop that has been bothering Mark for some time now.
    𝗦𝗽𝗲𝗮𝗸𝗲𝗿𝘀:
    * Scott Hanselman
    * Mark Russinovich
    𝗦𝗲𝘀𝘀𝗶𝗼𝗻 𝗜𝗻𝗳𝗼𝗿𝗺𝗮𝘁𝗶𝗼𝗻:
    This video is one of many sessions delivered for the Microsoft Build 2024 event. View the full session schedule and learn more about Microsoft Build at build.microsof...
    BRK255 | English (US) | Other
    #MSBuild

ความคิดเห็น • 22

  • @ashjaylk
    @ashjaylk 3 หลายเดือนก่อน +6

    man, i love this duo. Scott's full of presentation skills and Mark on the other hand super technical genius, still using his sysinternal tools

  • @NdamuleloNemakh
    @NdamuleloNemakh 4 หลายเดือนก่อน +8

    Always enjoy watching these two

  • @geehaf
    @geehaf 4 หลายเดือนก่อน +6

    Great stuff - fun to watch and learn new stuff at the same time. "..watching your Dad delete email..." classic. Love the C64 shirt too!

  • @charlesstacyharrisiii2593
    @charlesstacyharrisiii2593 4 หลายเดือนก่อน +8

    Fun session. Are you going to put the link to the source in the description?

  • @DontFretBrett
    @DontFretBrett 4 หลายเดือนก่อน +6

    Mark’s a little awkward but I liked the demo. Good job guys

  • @GeekyChristian
    @GeekyChristian 3 หลายเดือนก่อน +1

    Entertaining and very helpful. I also learned to program on the C64!

  • @tplummer217
    @tplummer217 4 หลายเดือนก่อน +1

    Cant wait to dig into this stuff

  • @PhilipMantom
    @PhilipMantom หลายเดือนก่อน

    @4:30 did Scott say "the world computer" ????

  • @BigEightiesNewWave
    @BigEightiesNewWave 4 หลายเดือนก่อน +6

    Mark, ages in reverse, and is a brain, and hunk.

  • @Bourn77
    @Bourn77 3 หลายเดือนก่อน +1

    please share the source code used for the demo, or a small sample semantic kernel app with local hosted model.

  • @YitzYBerlin
    @YitzYBerlin 9 วันที่ผ่านมา

    Are any of the txt files 0 bytes?

  • @ambroseleung4730
    @ambroseleung4730 4 หลายเดือนก่อน +1

    29:00 - How do you set up phi3 to run on localhost on the NPU? (I already know how to run SLMs locally with ollama, but it doesn't use NPU)

  • @sreesivs
    @sreesivs 4 หลายเดือนก่อน +1

    Superb...

  • @toulasantha
    @toulasantha 2 หลายเดือนก่อน

    Was that Chat GPT reply a record they played?😮

  • @midniteoilsoftware
    @midniteoilsoftware 3 หลายเดือนก่อน

    I have that same Commodore 64 t-shirt :)

  • @TakuCoding
    @TakuCoding 4 หลายเดือนก่อน +1

    This is really cool,

  • @mpabbi
    @mpabbi 3 หลายเดือนก่อน

    Source code?

  • @paulkeating9259
    @paulkeating9259 4 หลายเดือนก่อน +10

    Hi programmer here.... there is in fact 36 files..... 0 is counted, can't believe they let these 2 interns give the talk.

    • @chrisbaker5284
      @chrisbaker5284 3 หลายเดือนก่อน +2

      Do you mean 37 files but the bot counted 36 because it counted from 0? (off by one error).

    • @paulkeating9259
      @paulkeating9259 3 หลายเดือนก่อน

      @@chrisbaker5284 Sorry I am an AI chatbot and cannot answer such questions

  • @chrisbaker5284
    @chrisbaker5284 3 หลายเดือนก่อน

    Entertaining talk, but come on guys, 2 great minds in the computer industry and you couldn't figure out what the "hidden" text file was?!
    The stupid bot counted from 0, so it reported 36 files (despite it showing 37 files when it did the listing, one of you even said it was an off by 1 error, but then decided there was a hidden file 🤔).
    This doesn't bode well for the future of AI or the people that trust in it's accuracy.

    • @kevinmcfarlane2752
      @kevinmcfarlane2752 3 หลายเดือนก่อน

      In my experience of playing with these things over the past year I’ve found they are both amazingly impressive and surprisingly dumb. But eventually the dumb will be eliminated.
      However, I still think the imminence of AGI is an overestimate.