[
  {
    "start": 0.22,
    "end": 1.3,
    "text": "This podcast is"
  },
  {
    "start": 1.42,
    "end": 2.12,
    "text": "presented by"
  },
  {
    "start": 2.38,
    "end": 5.0,
    "text": "NXAI, your partner for time series."
  },
  {
    "start": 5.08,
    "end": 6.0,
    "text": "foundation models"
  },
  {
    "start": 6.1,
    "end": 7.06,
    "text": "and physical AI."
  },
  {
    "start": 19.06,
    "end": 19.92,
    "text": "Hello everybody!"
  },
  {
    "start": 19.98,
    "end": 23.22,
    "text": "And welcome to a new episode of our Industrial AI Podcast."
  },
  {
    "start": 23.32,
    "end": 25.18,
    "text": "my name's Robert Wieber and it's pleasure talk too"
  },
  {
    "start": 26.12,
    "end": 27.02,
    "text": "Peter Sieberg."
  },
  {
    "start": 27.14,
    "end": 32.9,
    "text": "good morning afternoon evening robert To all of you listeners wherever you are"
  },
  {
    "start": 33.38,
    "end": 33.76,
    "text": "peter."
  },
  {
    "start": 34.04,
    "end": 35.68,
    "text": "so A New Episode."
  },
  {
    "start": 35.76,
    "end": 36.7,
    "text": "what do You have?"
  },
  {
    "start": 37.88,
    "end": 39.7,
    "text": "Not that many things actually."
  },
  {
    "start": 40.12,
    "end": 47.66,
    "text": "I think you know, not always but sometimes on Sunday have a look at this arena leaderboard."
  },
  {
    "start": 47.76,
    "end": 51.18,
    "text": "so i did last sunday as we were recording yesterday."
  },
  {
    "start": 51.32,
    "end": 55.2,
    "text": "But for you listeners last sundae top ten and what is said?"
  },
  {
    "start": 55.34,
    "end": 56.58,
    "text": "open ai's back."
  },
  {
    "start": 56.76,
    "end": 59.32,
    "text": "they had left the first place to entropic."
  },
  {
    "start": 60.02,
    "end": 71.44,
    "text": "Now the Astra is second after entropic fable first and then there's a number of more fables another astra And then it's nine to ten and I'm gonna move into my topic here, right?"
  },
  {
    "start": 71.48,
    "end": 76.16,
    "text": "It is not just like looking at this but Nine in ten isn't for the open weight models."
  },
  {
    "start": 76.26,
    "end": 80.96,
    "text": "In this case moonshot Kimmy Tencent high four if that's how you pronounce it."
  },
  {
    "start": 81.58,
    "end": 94.18,
    "text": "My first question was another going to be a second more important question But the first one where as Google You know they have been out of their top ten for quite some time now independent shared every now and then."
  },
  {
    "start": 94.24,
    "end": 98.62,
    "text": "that I'm a happy private user of Gemini, but this doesn't help them."
  },
  {
    "start": 100.3,
    "end": 106.64,
    "text": "One in whatever two billion potential users they come in only at sixteen with their flash model."
  },
  {
    "start": 106.92,
    "end": 113.54,
    "text": "And again This is not about all these models But I'm just going to mention two, three more alongside in the top ten twenty."
  },
  {
    "start": 113.74,
    "end": 114.56,
    "text": "then there's meta."
  },
  {
    "start": 115.32,
    "end": 120.54,
    "text": "I think it's called Mu spark GLM from Z AI Quen Alibaba deep seek."
  },
  {
    "start": 120.9,
    "end": 121.78,
    "text": "they're all names."
  },
  {
    "start": 121.88,
    "end": 129.72,
    "text": "grok space X. some of you listeners will know all them by heart others do not and that is actually more."
  },
  {
    "start": 130.28,
    "end": 138.54,
    "text": "why i want to discuss this with you Robert also very specifically because It's more like do we care?"
  },
  {
    "start": 138.66,
    "end": 144.34,
    "text": "Do you care, do...do We need to know all of them in the top ten I mean In the end."
  },
  {
    "start": 144.44,
    "end": 145.26,
    "text": "Do we all wanna?"
  },
  {
    "start": 145.46,
    "end": 149.56,
    "text": "is the comparison right in The End and it may last at ten twenty years?"
  },
  {
    "start": 149.66,
    "end": 152.48,
    "text": "We in Germany would have said do we all want a drive Porsche?"
  },
  {
    "start": 152.62,
    "end": 155.66,
    "text": "can't we all allow afford Porsche, well we probably cannot."
  },
  {
    "start": 157.48,
    "end": 158.36,
    "text": "The need also."
  },
  {
    "start": 158.48,
    "end": 167.22,
    "text": "and then specifically or are you specifically users listeners more interested in your specific models for a specific requirements?"
  },
  {
    "start": 167.84,
    "end": 173.46,
    "text": "And I'm happy to call out XLSM as an example of let's say industrial time series right!"
  },
  {
    "start": 173.94,
    "end": 184.16,
    "text": "Then there is the second one and lets discuss it only after i was thinking about discussing that this thing opposed by BR, industrial automation."
  },
  {
    "start": 184.26,
    "end": 186.3,
    "text": "And they say AI is evolving fast."
  },
  {
    "start": 186.38,
    "end": 187.92,
    "text": "exactly why openness?"
  },
  {
    "start": 188.0,
    "end": 195.76,
    "text": "So there's a topic industry company should be able to benefit from the latest AI progress without being tied too close."
  },
  {
    "start": 196.34,
    "end": 197.48,
    "text": "so here you are."
  },
  {
    "start": 197.66,
    "end": 198.7,
    "text": "I don't go further."
  },
  {
    "start": 199.04,
    "end": 200.2,
    "text": "that's the topic of it."
  },
  {
    "start": 200.34,
    "end": 206.6,
    "text": "do our listeners need Do they need to follow the top ten?"
  },
  {
    "start": 207.02,
    "end": 208.08,
    "text": "What's happening there?"
  },
  {
    "start": 208.56,
    "end": 208.94,
    "text": "or will."
  },
  {
    "start": 209.02,
    "end": 209.8,
    "text": "They have chosen."
  },
  {
    "start": 209.88,
    "end": 210.9,
    "text": "what is your experience?"
  },
  {
    "start": 210.96,
    "end": 218.68,
    "text": "you are out here a lot more these days and I am Robert, Have they already chose in the majority to go open should be able too."
  },
  {
    "start": 218.96,
    "end": 226.46,
    "text": "if for whatever reason that views LLM A four half year anyone has switched to B Should That Be Possible?"
  },
  {
    "start": 226.58,
    "end": 227.4,
    "text": "Is That Possible?"
  },
  {
    "start": 227.9,
    "end": 229.6,
    "text": "What Does Your Feeling Your Experience?"
  },
  {
    "start": 230.06,
    "end": 233.44,
    "text": "So first of all i ever answer where Gemini is."
  },
  {
    "start": 233.88,
    "end": 238.82,
    "text": "So Gemini is running in my car, so it was I think with the latest update."
  },
  {
    "start": 238.94,
    "end": 247.74,
    "text": "I got a Gemini in the car because i have an Android based car system and they run Gemini there..."
  },
  {
    "start": 248.2,
    "end": 249.6,
    "text": "It's different voice by the way!"
  },
  {
    "start": 249.7,
    "end": 251.0,
    "text": "Because I use both of them."
  },
  {
    "start": 251.56,
    "end": 254.22,
    "text": "It's a different system from one you do on your hand."
  },
  {
    "start": 255.72,
    "end": 257.6,
    "text": "Exactly that's our first answer."
  },
  {
    "start": 259.04,
    "end": 277.38,
    "text": "I hear in the moment that a lot of industry companies are working with quen models because it's an open environment and everybody who is doing something when it comes to engineering coding stuff, they always offer all kinds of model."
  },
  {
    "start": 277.5,
    "end": 278.92,
    "text": "so you can choose your model."
  },
  {
    "start": 279.02,
    "end": 280.7,
    "text": "Your favorite model You can have?"
  },
  {
    "start": 280.78,
    "end": 282.32,
    "text": "no may I?"
  },
  {
    "start": 285.96,
    "end": 289.7,
    "text": "Everybody wants to offer an open environment."
  },
  {
    "start": 289.82,
    "end": 309.02,
    "text": "To use your specialized model and what I also heard last week is from a Chinese company that they don't want to run their system on deep-seek models because there's so much influence from the Chinese government or Chinese Company, And they are now also switching to quen models."
  },
  {
    "start": 309.46,
    "end": 310.76,
    "text": "So that's interesting."
  },
  {
    "start": 311.46,
    "end": 312.5,
    "text": "But isn't, isn't quen"
  },
  {
    "start": 313.36,
    "end": 313.6,
    "text": "also?"
  },
  {
    "start": 315.08,
    "end": 315.2,
    "text": "Yeah."
  },
  {
    "start": 315.24,
    "end": 322.34,
    "text": "It's all China based but I'm not sure where they're changing because deep seek is not the favorite anymore in this company."
  },
  {
    "start": 323.6,
    "end": 327.62,
    "text": "Right, so there's a couple of topics here because you said and specifically also."
  },
  {
    "start": 327.7,
    "end": 330.66,
    "text": "You mentioned an example over closed on open."
  },
  {
    "start": 330.78,
    "end": 338.16,
    "text": "So if these Solution providers give you the option to choose whatever you want to choose And maybe switch wherever you wanna choose?"
  },
  {
    "start": 338.62,
    "end": 345.14,
    "text": "You can they offer you then also in the option of going for a close or open-opened weights model"
  },
  {
    "start": 346.22,
    "end": 346.64,
    "text": "exactly."
  },
  {
    "start": 346.9,
    "end": 348.4,
    "text": "They leave it up to you completely done."
  },
  {
    "start": 348.78,
    "end": 354.24,
    "text": "now I'm not sure if you have so you mention are almost like a change going from one to the other."
  },
  {
    "start": 354.32,
    "end": 358.76,
    "text": "That's the other interesting piece that I do not have, experience."
  },
  {
    "start": 358.84,
    "end": 359.94,
    "text": "maybe you've heard something."
  },
  {
    "start": 360.02,
    "end": 360.8,
    "text": "how easy is it?"
  },
  {
    "start": 360.86,
    "end": 362.4,
    "text": "To move for want of your model?"
  },
  {
    "start": 362.5,
    "end": 370.1,
    "text": "It's like can i just switch From A to B and expect the type of answers, that type of work."
  },
  {
    "start": 370.18,
    "end": 375.7,
    "text": "I'm gonna have a piece later on Fabian and can i switch from a worker A to B?"
  },
  {
    "start": 375.96,
    "end": 376.0,
    "text": "No!"
  },
  {
    "start": 376.18,
    "end": 376.7,
    "text": "I cannot."
  },
  {
    "start": 376.8,
    "end": 382.82,
    "text": "you know if one person leaves the company uh am not going to have any new person... That takes it takes off for years."
  },
  {
    "start": 383.32,
    "end": 386.92,
    "text": "so I'm going to ask in that way do you- Have You Heard Of This?"
  },
  {
    "start": 387.14,
    "end": 389.82,
    "text": "Do People Say Within Twenty Four Hours Or Whatever?"
  },
  {
    "start": 390.1,
    "end": 390.6,
    "text": "No Do"
  },
  {
    "start": 390.72,
    "end": 405.08,
    "text": "You Remember The Presentation By Reiner Brehm one and a half year ago when he described how they build the first Siemens agentic solution or LLM shop floor, LLm stuff like that."
  },
  {
    "start": 405.58,
    "end": 412.2,
    "text": "And then over the night Microsoft changed the weights of the model and the answers went crazy."
  },
  {
    "start": 412.82,
    "end": 420.48,
    "text": "I think there are not so much problems anymore When it comes to that because i will record an episode with A huge automation company."
  },
  {
    "start": 420.98,
    "end": 425.88,
    "text": "we Will offer an open environment for models like this."
  },
  {
    "start": 426.46,
    "end": 436.8,
    "text": "I'm not allowed to talk about now, because they will present it at SPS but i'll ask them what does that mean?"
  },
  {
    "start": 436.9,
    "end": 442.42,
    "text": "to a Czech GPT or to a Gemini, or through a Grog something."
  },
  {
    "start": 442.58,
    "end": 445.0,
    "text": "Very good very important and much looking forward too."
  },
  {
    "start": 445.08,
    "end": 446.58,
    "text": "because let's stay with the cars."
  },
  {
    "start": 446.64,
    "end": 448.7,
    "text": "I mean everybody knows and understands."
  },
  {
    "start": 448.8,
    "end": 465.06,
    "text": "there is at least when The only thing that Is really different even internationally That if we move to another country Europeans going To United Kingdom Or countries I believe Singapore others many other Countries it's left or right driving."
  },
  {
    "start": 465.12,
    "end": 468.9,
    "text": "but independent of that you know driving has been so far."
  },
  {
    "start": 470.26,
    "end": 478.84,
    "text": "We're making it easier by moving to autonomous driving, but I mean let's stay tuned as a wheel on there is we call the gas for acceleration does a break etc."
  },
  {
    "start": 479.2,
    "end": 495.54,
    "text": "that's... It almost like its been normed and thats why you can just jump in from a Volkswagen into Porsche or Audi whatever independent of if its diesel or gas or in the meantime battery driven."
  },
  {
    "start": 497.36,
    "end": 505.44,
    "text": "So maybe at some point in time we can make this comparison would be will be great if it's some point of Time that is as easy as there, but I'm not sure."
  },
  {
    "start": 505.58,
    "end": 508.34,
    "text": "It is already As easy as that today."
  },
  {
    "start": 508.42,
    "end": 510.74,
    "text": "so very much looking forward to the answer That you're gonna get."
  },
  {
    "start": 510.88,
    "end": 510.98,
    "text": "yeah"
  },
  {
    "start": 511.5,
    "end": 511.88,
    "text": "perfect."
  },
  {
    "start": 512.7,
    "end": 525.36,
    "text": "i've brought an interesting question with me peter because last week i met Stratus from University Of Amsterdam during a client meeting and That evening he asked Me A Question That stuck With Me The whole Week."
  },
  {
    "start": 526.32,
    "end": 529.34,
    "text": "in what modality do you think, Peter?"
  },
  {
    "start": 530.26,
    "end": 530.36,
    "text": "Yeah."
  },
  {
    "start": 531.14,
    "end": 534.64,
    "text": "Language text images or while else is there?"
  },
  {
    "start": 535.42,
    "end": 537.04,
    "text": "What modality Do You use?"
  },
  {
    "start": 537.98,
    "end": 540.6,
    "text": "and I think yeah as a person"
  },
  {
    "start": 540.66,
    "end": 541.64,
    "text": "depending on who i am."
  },
  {
    "start": 542.12,
    "end": 544.68,
    "text": "what does my use case what Is My Persona right here?"
  },
  {
    "start": 544.82,
    "end": 545.84,
    "text": "your you are a person."
  },
  {
    "start": 546.12,
    "end": 546.68,
    "text": "you're human."
  },
  {
    "start": 547.18,
    "end": 556.86,
    "text": "what Does Your Modality to Think And I have been thinking about it ever since because it's not language."
  },
  {
    "start": 557.54,
    "end": 559.68,
    "text": "It's not image, it's more thought."
  },
  {
    "start": 559.88,
    "end": 563.34,
    "text": "there is a missing topic right says the missing missing link."
  },
  {
    "start": 563.42,
    "end": 564.44,
    "text": "what does your opinion on that?"
  },
  {
    "start": 565.14,
    "end": 566.86,
    "text": "Yeah sure okay well Okay."
  },
  {
    "start": 567.12,
    "end": 577.78,
    "text": "so at The end of your piece I was thinking about closing my eyes and then one modality has gone because My answer would be immediately language."
  },
  {
    "start": 578.04,
    "end": 581.14,
    "text": "i believe languages the way mean just sticking to."
  },
  {
    "start": 581.74,
    "end": 582.58,
    "text": "That's my priority."
  },
  {
    "start": 582.62,
    "end": 583.4,
    "text": "Is that enough?"
  },
  {
    "start": 583.68,
    "end": 584.38,
    "text": "It is language?"
  },
  {
    "start": 584.7,
    "end": 593.74,
    "text": "I'm not sure, so i'm just saying that this might current if in the past and it was more like a search wanting to know things."
  },
  {
    "start": 593.8,
    "end": 596.4,
    "text": "but at the end shop floor person also wants you now."
  },
  {
    "start": 596.74,
    "end": 598.28,
    "text": "why are they still working?"
  },
  {
    "start": 598.72,
    "end": 600.78,
    "text": "why is the line stopping actually even?"
  },
  {
    "start": 600.96,
    "end": 607.98,
    "text": "of course we're gonna get the answer before These go."
  },
  {
    "start": 609.64,
    "end": 615.48,
    "text": "I said it in a recording couple of days ago."
  },
  {
    "start": 616.02,
    "end": 620.26,
    "text": "sometimes i tell the complete story like i do you know yes, why no?"
  },
  {
    "start": 620.8,
    "end": 625.8,
    "text": "but by telling this story which is then maybe only like four or five sentences."
  },
  {
    "start": 626.44,
    "end": 633.72,
    "text": "My feeling is that I get a very high quality, at least the typing isn't no way for me."
  },
  {
    "start": 633.86,
    "end": 635.8,
    "text": "That's old and whole school right?"
  },
  {
    "start": 636.56,
    "end": 649.02,
    "text": "And then to finish it you said well maybe its not language but whatever happens around favorite movie again yesterday and it's like Night on Earth, right?"
  },
  {
    "start": 649.16,
    "end": 650.32,
    "text": "By Jim Jarmusch."
  },
  {
    "start": 650.94,
    "end": 656.2,
    "text": "And for the people who have seen it under third sequence is The Blind Lady in a Taxi In Paris."
  },
  {
    "start": 656.88,
    "end": 659.46,
    "text": "so she feels things."
  },
  {
    "start": 659.54,
    "end": 664.86,
    "text": "but now we normal what is normal...the People Who Can See We Don't Feel Things."
  },
  {
    "start": 664.92,
    "end": 666.64,
    "text": "So maybe that Is What You're Suggesting."
  },
  {
    "start": 666.68,
    "end": 668.16,
    "text": "That Is The Environment Around Us."
  },
  {
    "start": 668.64,
    "end": 673.32,
    "text": "Yeah Maybe Language Is Only The Output Of Thinking Right."
  },
  {
    "start": 673.9,
    "end": 676.7,
    "text": "So when you talk to someone, it's the output."
  },
  {
    "start": 677.2,
    "end": 680.1,
    "text": "But thinking was before what we would tell that person."
  },
  {
    "start": 680.5,
    "end": 691.7,
    "text": "so It wasn't a very interesting discussion because Stratus said yeah Language for sure image for sure but there must be even more sensing sensors haptic feedback and stuff like That."
  },
  {
    "start": 692.32,
    "end": 699.84,
    "text": "And he raised The question Because if We want To look Beyond large language Model and Consider For example Robotics?"
  },
  {
    "start": 702.38,
    "end": 703.92,
    "text": "what thinking means, right?"
  },
  {
    "start": 704.08,
    "end": 708.78,
    "text": "Not only language image sensing haptical feedback."
  },
  {
    "start": 709.04,
    "end": 714.06,
    "text": "What else is there in this whole topic when it comes to thinking so interesting question?"
  },
  {
    "start": 714.12,
    "end": 715.66,
    "text": "maybe perhaps you do."
  },
  {
    "start": 715.76,
    "end": 719.76,
    "text": "listeners are going the same way now and think about this question the whole week."
  },
  {
    "start": 721.94,
    "end": 727.1,
    "text": "but I like to share The Question by Stratus because i think its a very important question for all of us."
  },
  {
    "start": 727.62,
    "end": 727.98,
    "text": "Yeah."
  },
  {
    "start": 728.48,
    "end": 733.2,
    "text": "And if you want to discuss with stars directly, I think he's going to be with us in so what?"
  },
  {
    "start": 733.28,
    "end": 734.9,
    "text": "we'll come back to that a little bit later."
  },
  {
    "start": 734.94,
    "end": 740.26,
    "text": "i'm gonna share it up but He's Gonna Be at our AI at the volcano as well."
  },
  {
    "start": 740.3,
    "end": 741.46,
    "text": "on the volcano on"
  },
  {
    "start": 741.86,
    "end": 747.9,
    "text": "the Volcano There are still A couple of Seats left So You Would Still Have The Opportunity To Discuss With Him Directly."
  },
  {
    "start": 747.94,
    "end": 748.08,
    "text": "yeah"
  },
  {
    "start": 748.14,
    "end": 748.3,
    "text": "i Think."
  },
  {
    "start": 748.4,
    "end": 748.98,
    "text": "Two"
  },
  {
    "start": 749.44,
    "end": 750.32,
    "text": "Very Good Question."
  },
  {
    "start": 750.42,
    "end": 756.86,
    "text": "It Feels Also Little Bit That You're Moving Forget His Name All The Time Our French Friend Who'S also moving?"
  },
  {
    "start": 757.12,
    "end": 762.0,
    "text": "I mean, his direct answer is it's not language."
  },
  {
    "start": 762.04,
    "end": 763.28,
    "text": "He always... What's his name?"
  },
  {
    "start": 764.42,
    "end": 765.88,
    "text": "Jan has been saying that."
  },
  {
    "start": 765.96,
    "end": 770.94,
    "text": "right It cannot be and thats why he started this company to go into the world."
  },
  {
    "start": 771.16,
    "end": 777.94,
    "text": "i assume which Stratus again talking discussing with us about World models Is that?"
  },
  {
    "start": 778.32,
    "end": 781.46,
    "text": "No, he's talking about how to deploy physical AI."
  },
  {
    "start": 781.58,
    "end": 783.88,
    "text": "So it's more about the deployment topic this time."
  },
  {
    "start": 784.02,
    "end": 785.32,
    "text": "Okay very good."
  },
  {
    "start": 785.4,
    "end": 787.1,
    "text": "so yeah let's discuss a little bit more later."
  },
  {
    "start": 787.2,
    "end": 790.38,
    "text": "I think language is important but there are even more."
  },
  {
    "start": 790.88,
    "end": 794.86,
    "text": "and then we talked about memory different kinds of memory."
  },
  {
    "start": 795.92,
    "end": 800.74,
    "text": "It will be interesting in October with Stratis In the evening with beer."
  },
  {
    "start": 803.14,
    "end": 804.86,
    "text": "You see beer is also important."
  },
  {
    "start": 807.86,
    "end": 811.06,
    "text": "Let me share the one I already refer to according too."
  },
  {
    "start": 811.28,
    "end": 812.74,
    "text": "We are waiting for you a book."
  },
  {
    "start": 812.9,
    "end": 814.2,
    "text": "Fabian court Maya."
  },
  {
    "start": 814.56,
    "end": 818.94,
    "text": "Yes, AI agents need onboarding same as with humans."
  },
  {
    "start": 819.24,
    "end": 823.14,
    "text": "so new colleague He says is provided with the best practices."
  },
  {
    "start": 823.22,
    "end": 825.58,
    "text": "videos may be specialized training whatever."
  },
  {
    "start": 825.7,
    "end": 832.36,
    "text": "same way AI Agents should receive context tool descriptions examples Constraints."
  },
  {
    "start": 832.48,
    "end": 840.06,
    "text": "it needs to operate within this specific workflow and of course MCP already provides a way to expose the tools."
  },
  {
    "start": 840.2,
    "end": 855.68,
    "text": "There's a little bit there, I'm not gonna go further details but what i understood and referred with before is like...I think he saying it's almost treat your agents same way you treat new human colleagues."
  },
  {
    "start": 856.1,
    "end": 857.12,
    "text": "they need onboarding."
  },
  {
    "start": 857.5,
    "end": 860.24,
    "text": "do not just send the agents."
  },
  {
    "start": 861.34,
    "end": 867.04,
    "text": "have them go out doing things for you without Telling them who you are."
  },
  {
    "start": 867.2,
    "end": 867.84,
    "text": "You know, Who?"
  },
  {
    "start": 867.98,
    "end": 870.94,
    "text": "Are you as a company in what department or your working?"
  },
  {
    "start": 871.2,
    "end": 873.1,
    "text": "What is the specific thing that you want to?"
  },
  {
    "start": 873.28,
    "end": 883.88,
    "text": "so I think it's the same with humans and maybe It's almost like close through The topic we just had but its Like if you do not Know where you are And Where he sitting."
  },
  {
    "start": 883.96,
    "end": 886.42,
    "text": "Is this private This business?"
  },
  {
    "start": 886.94,
    "end": 887.86,
    "text": "what is the environment?"
  },
  {
    "start": 888.2,
    "end": 892.32,
    "text": "What does my boss want me to do, I as a human etc."
  },
  {
    "start": 893.52,
    "end": 901.42,
    "text": "It's great question and it has been thought that he is considering making sure you onboard your agents."
  },
  {
    "start": 901.94,
    "end": 902.36,
    "text": "Perfect!"
  },
  {
    "start": 903.06,
    "end": 916.54,
    "text": "And i need to share something because we are proud of our colleagues at NVIDIA who shared their podcast with their community And that really means something, and it also shows in the download numbers too."
  },
  {
    "start": 916.74,
    "end": 920.36,
    "text": "So you can do that again, NVIDIA!"
  },
  {
    "start": 920.92,
    "end": 921.88,
    "text": "That helps us a lot."
  },
  {
    "start": 922.46,
    "end": 925.66,
    "text": "but yeah we are bit proud because its not the daily..."
  },
  {
    "start": 926.12,
    "end": 929.58,
    "text": "It's the one where you recorded about safety right?"
  },
  {
    "start": 929.98,
    "end": 931.34,
    "text": "Yeah it is huge topic for NVIDia."
  },
  {
    "start": 931.46,
    "end": 932.72,
    "text": "I think so interesting."
  },
  {
    "start": 937.1,
    "end": 937.76,
    "text": "to contact."
  },
  {
    "start": 938.02,
    "end": 938.66,
    "text": "What was the name?"
  },
  {
    "start": 938.7,
    "end": 939.6,
    "text": "Was it Massimo or..."
  },
  {
    "start": 940.04,
    "end": 948.08,
    "text": "It's was Ricardo Mariani and he, I think it was Tim Fisher is a part of our AI group."
  },
  {
    "start": 948.7,
    "end": 958.48,
    "text": "I met him at SICK and he said yes you should talk to NVIDIA because they are doing something when it comes to safety And so there were some connections."
  },
  {
    "start": 958.76,
    "end": 961.94,
    "text": "thanks a lot for this connection with NVIDia"
  },
  {
    "start": 962.82,
    "end": 964.12,
    "text": "And the takeaway message."
  },
  {
    "start": 964.18,
    "end": 977.98,
    "text": "as far as I recall, not sure."
  },
  {
    "start": 990.38,
    "end": 994.78,
    "text": "new approach coming from the automotive industry and now going into the factories."
  },
  {
    "start": 994.86,
    "end": 995.52,
    "text": "very interesting."
  },
  {
    "start": 995.58,
    "end": 999.4,
    "text": "And yeah, he always underlines opportunity of safety."
  },
  {
    "start": 999.56,
    "end": 1000.62,
    "text": "I would say Yeah"
  },
  {
    "start": 1000.88,
    "end": 1004.98,
    "text": "exactly well good luck or congratulations to you."
  },
  {
    "start": 1005.54,
    "end": 1007.02,
    "text": "No not too us Peter."
  },
  {
    "start": 1010.36,
    "end": 1010.88,
    "text": "You chose it."
  },
  {
    "start": 1010.92,
    "end": 1014.02,
    "text": "then you picked that specific podcast."
  },
  {
    "start": 1014.32,
    "end": 1015.12,
    "text": "So what else do we have?"
  },
  {
    "start": 1015.76,
    "end": 1019.42,
    "text": "Well, I have a closing one which goes towards Weetron then."
  },
  {
    "start": 1019.52,
    "end": 1034.94,
    "text": "So Weeton is the company actually providing automated logistics solutions for the food retail industry and they have an order picking network OPN for rethinking logistics right."
  },
  {
    "start": 1035.04,
    "end": 1045.18,
    "text": "so instead of what they say optimizing individual warehouse locations and maybe you can talk more about it because you know the very well Robert or processes in isolation."
  },
  {
    "start": 1045.44,
    "end": 1049.44,
    "text": "Maybe we talked about this for a topic I now think may be like four years ago."
  },
  {
    "start": 1049.68,
    "end": 1051.08,
    "text": "i think, how will tell you?"
  },
  {
    "start": 1051.34,
    "end": 1052.72,
    "text": "are we wrote about that in our book?"
  },
  {
    "start": 1052.84,
    "end": 1053.5,
    "text": "i think so."
  },
  {
    "start": 1053.66,
    "end": 1060.72,
    "text": "they viewed entire value chain as an interconnected system from the retail store transportation what do i think?"
  },
  {
    "start": 1060.82,
    "end": 1063.32,
    "text": "but not you gave me example at that time."
  },
  {
    "start": 1063.52,
    "end": 1075.96,
    "text": "you told me you said If there is going to be a traffic jam for a truck, truck delivering whatever full of food and they would know in the distribution center."
  },
  {
    "start": 1076.08,
    "end": 1082.36,
    "text": "In the end they will know it in their retails store, which orders from the distribution center that this you know."
  },
  {
    "start": 1082.48,
    "end": 1087.7,
    "text": "think about south of where I live on a Monday typically moving into Austria."
  },
  {
    "start": 1087.78,
    "end": 1092.72,
    "text": "That's going to be probably ten thousand trucks waiting to be allowed To move over."
  },
  {
    "start": 1092.8,
    "end": 1095.88,
    "text": "so if that's what you know maybe in general or something is happening?"
  },
  {
    "start": 1098.4,
    "end": 1104.66,
    "text": "We can talk about a little bit more, but the point also here was and you can listen to a podcast."
  },
  {
    "start": 1104.7,
    "end": 1114.5,
    "text": "Maybe want to talk about that with their vitron CEOs Helmut Breschenk Karl Högen discuss why OPN represents true paradigm shift."
  },
  {
    "start": 1115.1,
    "end": 1117.88,
    "text": "Yeah, it's a very interesting approach not the heart."
  },
  {
    "start": 1117.92,
    "end": 1124.76,
    "text": "A very horizontal view over the whole supply chain and AI I think also plays their crucial role."
  },
  {
    "start": 1124.84,
    "end": 1136.72,
    "text": "maybe It's not LLM driven its more tabular data time series data And stuff like that because you have different silos in the whole supply chain, right?"
  },
  {
    "start": 1136.78,
    "end": 1138.32,
    "text": "The transportation company."
  },
  {
    "start": 1138.56,
    "end": 1142.02,
    "text": "The warehouse, the retail store or farmer."
  },
  {
    "start": 1142.44,
    "end": 1150.24,
    "text": "so they all their data and main topic is really to crack these data silos to work with this data."
  },
  {
    "start": 1150.8,
    "end": 1153.78,
    "text": "And yeah, it's very interesting approach is a very interesting episode."
  },
  {
    "start": 1153.92,
    "end": 1168.1,
    "text": "I think they are already running some experiments was their customer in some countries and i think that will be really interesting discussion at our event because the sea level or the main players of each one also taking part in all groups."
  },
  {
    "start": 1168.26,
    "end": 1172.52,
    "text": "end thing its very interesting out so to discuss their chopstick there."
  },
  {
    "start": 1173.46,
    "end": 1173.94,
    "text": "Yeah, thanks."
  },
  {
    "start": 1173.98,
    "end": 1176.9,
    "text": "That was of course the second reason for sharing this."
  },
  {
    "start": 1176.98,
    "end": 1180.62,
    "text": "so first listen to our podcast and then Listen to that."
  },
  {
    "start": 1180.7,
    "end": 1181.92,
    "text": "we don't podcast."
  },
  {
    "start": 1182.38,
    "end": 1198.06,
    "text": "And if you say oh that's a really interesting topic which I really believe it is Dan You actually still have the opportunity to join Robert We have only We don't have to say too much about it."
  },
  {
    "start": 1198.18,
    "end": 1206.6,
    "text": "Anyway, for those of you that maybe do not know we're doing this every now and then... ...we've got our big AI in the Alps,... ....and a second one in Autumn."
  },
  {
    "start": 1206.66,
    "end": 1211.04,
    "text": "And at this time I may be wanting to share because you are so close with them."
  },
  {
    "start": 1211.08,
    "end": 1211.1,
    "text": "No!"
  },
  {
    "start": 1211.14,
    "end": 1212.4,
    "text": "You could- We call it A.I."
  },
  {
    "start": 1212.6,
    "end": 1213.54,
    "text": "At The Volcano?"
  },
  {
    "start": 1213.74,
    "end": 1213.98,
    "text": "On The"
  },
  {
    "start": 1214.14,
    "end": 1215.42,
    "text": "Volcano Peter!"
  },
  {
    "start": 1215.46,
    "end": 1217.32,
    "text": "On The volcano!"
  },
  {
    "start": 1217.72,
    "end": 1218.68,
    "text": "Why did we called there?"
  },
  {
    "start": 1218.72,
    "end": 1219.44,
    "text": "Do We Want To Light"
  },
  {
    "start": 1219.5,
    "end": 1219.6,
    "text": "It?,"
  },
  {
    "start": 1219.86,
    "end": 1220.6,
    "text": "Do We Wanna Share"
  },
  {
    "start": 1220.64,
    "end": 1220.66,
    "text": "it?..."
  },
  {
    "start": 1220.72,
    "end": 1222.44,
    "text": "...why would they call us like that or not?"
  },
  {
    "start": 1222.84,
    "end": 1223.56,
    "text": "Now you should join."
  },
  {
    "start": 1226.0,
    "end": 1228.08,
    "text": "So go for the two seats left."
  },
  {
    "start": 1228.68,
    "end": 1233.98,
    "text": "October, twenty eight thirty in Parkstein that's in Bavaria In Germany."
  },
  {
    "start": 1234.7,
    "end": 1236.08,
    "text": "we're going to be talking about an upper"
  },
  {
    "start": 1236.16,
    "end": 1237.04,
    "text": "plutonium."
  },
  {
    "start": 1237.12,
    "end": 1238.58,
    "text": "Oh is that English?"
  },
  {
    "start": 1242.18,
    "end": 1242.58,
    "text": "Topics"
  },
  {
    "start": 1242.74,
    "end": 1242.86,
    "text": "are."
  },
  {
    "start": 1242.94,
    "end": 1249.92,
    "text": "We just talked about robotics stratus gravis from Amsterdam You know and you met him last week."
  },
  {
    "start": 1249.96,
    "end": 1250.28,
    "text": "he was gonna."
  },
  {
    "start": 1251.36,
    "end": 1254.18,
    "text": "I just did a podcast with Bastian Al from Simcon."
  },
  {
    "start": 1254.4,
    "end": 1259.42,
    "text": "He is very interesting in simulations, Captain Marco Huber from Fraunhofer."
  },
  {
    "start": 1259.58,
    "end": 1263.9,
    "text": "he always does our use cases and not sure we still have another one for you"
  },
  {
    "start": 1263.96,
    "end": 1266.24,
    "text": "would have a genetic approach but"
  },
  {
    "start": 1270.86,
    "end": 1274.88,
    "text": "And maybe even a tour of one of the latest distribution centers."
  },
  {
    "start": 1275.18,
    "end": 1277.62,
    "text": "Absolutely, we are looking forward to do that."
  },
  {
    "start": 1278.26,
    "end": 1282.86,
    "text": "and now we switch to the main part because Alexander from Dunya is waiting."
  },
  {
    "start": 1283.56,
    "end": 1287.54,
    "text": "like so many others he's working on AI for new industrial materials."
  },
  {
    "start": 1287.7,
    "end": 1294.74,
    "text": "however explains what does differently compared to Max Welling, our entire pick or all the others."
  },
  {
    "start": 1295.22,
    "end": 1302.42,
    "text": "It's a Berlin based company very interesting approach and I think it's worth to listen to Alexander."
  },
  {
    "start": 1303.72,
    "end": 1305.6,
    "text": "Very good looking forward to Robert."
  },
  {
    "start": 1305.66,
    "end": 1307.66,
    "text": "Goodluck To you with your presentation."
  },
  {
    "start": 1307.72,
    "end": 1310.1,
    "text": "by the time that you hear this they'll listen to."
  },
  {
    "start": 1310.2,
    "end": 1316.9,
    "text": "Robert will have spoken And then i'm sure You're gonna share All about how it was at the Agilox Agilocks"
  },
  {
    "start": 1317.14,
    "end": 1317.98,
    "text": "Tech Talks?"
  },
  {
    "start": 1318.58,
    "end": 1320.06,
    "text": "I would start with Stratus question"
  },
  {
    "start": 1323.0,
    "end": 1326.88,
    "text": "And now you're looking forward to the people in the audience sharing their thoughts with you."
  },
  {
    "start": 1327.12,
    "end": 1327.5,
    "text": "Exactly,"
  },
  {
    "start": 1327.9,
    "end": 1328.84,
    "text": "Peter thanks a lot"
  },
  {
    "start": 1329.24,
    "end": 1330.04,
    "text": "Robert take care."
  },
  {
    "start": 1330.16,
    "end": 1331.1,
    "text": "greetings to Munich."
  },
  {
    "start": 1331.38,
    "end": 1339.02,
    "text": "bye-bye Hello everybody and welcome to new episode of our industrial AI podcast."
  },
  {
    "start": 1339.08,
    "end": 1339.74,
    "text": "my name is Hubert Weber."
  },
  {
    "start": 1340.34,
    "end": 1348.26,
    "text": "I'm recording on A parking lot next To The Highway Of The Beautiful City Of Passau and My guest Is based In Berlin Alexander Hammer."
  },
  {
    "start": 1348.34,
    "end": 1349.3,
    "text": "Welcome To The Podcast."
  },
  {
    "start": 1349.86,
    "end": 1350.58,
    "text": "Thanks a lot, Robert."
  },
  {
    "start": 1351.06,
    "end": 1354.68,
    "text": "You are the founder CEO co-founder of Dunia."
  },
  {
    "start": 1355.06,
    "end": 1356.36,
    "text": "What does dunia do?"
  },
  {
    "start": 1356.52,
    "end": 1359.3,
    "text": "Please introduce yourself and then dunia."
  },
  {
    "start": 1359.8,
    "end": 1360.46,
    "text": "so I'm Alex."
  },
  {
    "start": 1360.72,
    "end": 1363.28,
    "text": "i'm one Of three founders of dunia."
  },
  {
    "start": 1363.5,
    "end": 1366.78,
    "text": "dunia as you mentioned is a berlin based startup company."
  },
  {
    "start": 1367.3,
    "end": 1373.14,
    "text": "we use AI And lab automation to develop next generation materials."
  },
  {
    "start": 1373.78,
    "end": 1377.84,
    "text": "Okay So ai help to develop new material right"
  },
  {
    "start": 1378.24,
    "end": 1378.58,
    "text": "correct."
  },
  {
    "start": 1379.52,
    "end": 1381.92,
    "text": "How does AI help to develop new materials?"
  },
  {
    "start": 1382.04,
    "end": 1383.86,
    "text": "Can you explain that a little bit to us."
  },
  {
    "start": 1384.28,
    "end": 1384.8,
    "text": "Yeah, sure!"
  },
  {
    "start": 1385.5,
    "end": 1400.58,
    "text": "So what AI already does really well is it basically helps accelerate simulations and we can even use generative AI models purpose-built for materials To come up with some potential candidates on the computer screen."
  },
  {
    "start": 1400.94,
    "end": 1402.14,
    "text": "so just using simulations."
  },
  {
    "start": 1402.22,
    "end": 1408.34,
    "text": "Basically now with machine learning We could really speed up What we used do I guess traditional simulation approaches"
  },
  {
    "start": 1409.28,
    "end": 1409.64,
    "text": "and"
  },
  {
    "start": 1410.08,
    "end": 1411.92,
    "text": "that has actually worked really well in the last five years."
  },
  {
    "start": 1411.96,
    "end": 1413.78,
    "text": "There's been a lot of progress over there."
  },
  {
    "start": 1414.18,
    "end": 1419.28,
    "text": "The other big push is building what people call an AI scientist."
  },
  {
    "start": 1420.22,
    "end": 1422.96,
    "text": "You've seen that LMs got really good at writing emails."
  },
  {
    "start": 1423.1,
    "end": 1425.72,
    "text": "They also get really good and generating images, right?"
  },
  {
    "start": 1425.76,
    "end": 1433.0,
    "text": "So increasingly we're learning how to generate novel compositions of matter by putting atoms together in unique ways."
  },
  {
    "start": 1433.16,
    "end": 1442.08,
    "text": "There's still a few issues there but I think between LM-based AI scientists doing what the human research does entities are much more powerful simulation engines."
  },
  {
    "start": 1442.88,
    "end": 1448.62,
    "text": "People start using AI now for their physical world very foundation on everything material science."
  },
  {
    "start": 1449.08,
    "end": 1453.52,
    "text": "When we dig a little bit deeper before we do that, please introduce yourself briefly to the listeners."
  },
  {
    "start": 1453.92,
    "end": 1457.56,
    "text": "Are you an AI guy or are your domain specific?"
  },
  {
    "start": 1457.98,
    "end": 1459.94,
    "text": "Guy coming from a specific domain?"
  },
  {
    "start": 1460.12,
    "end": 1462.78,
    "text": "so I did a PhD in chem informatics."
  },
  {
    "start": 1462.9,
    "end": 1463.44,
    "text": "right now"
  },
  {
    "start": 1463.56,
    "end": 1464.46,
    "text": "That's interesting cute"
  },
  {
    "start": 1464.5,
    "end": 1464.92,
    "text": "designs."
  },
  {
    "start": 1464.96,
    "end": 1468.32,
    "text": "Yeah very much interdisciplinary and away for my background."
  },
  {
    "start": 1468.82,
    "end": 1484.04,
    "text": "having said class like a very traditional chemist right when I was uh eighteen years old, i wanted to cure cancer and I got really interested in organic chemistry how to put molecules together to make potential drug molecules."
  },
  {
    "start": 1484.5,
    "end": 1487.88,
    "text": "So that's where I started also very much as an experimentalist."
  },
  {
    "start": 1488.02,
    "end": 1491.58,
    "text": "so I stood up a lot in labs Also did the same thing in industry."
  },
  {
    "start": 1492.42,
    "end": 1496.26,
    "text": "oh I saw how that worked And then there were few things that happened."
  },
  {
    "start": 1496.48,
    "end": 1500.64,
    "text": "um That caught me quite early quite excited about what machine learning can do."
  },
  {
    "start": 1501.14,
    "end": 1505.96,
    "text": "And then over the years, I increasingly started working more with phone computers to make my job easier."
  },
  {
    "start": 1511.0,
    "end": 1517.66,
    "text": "and then now it's been almost ten years that I've been working on basically what ultimately became Dunja, but we're still doing today."
  },
  {
    "start": 1517.96,
    "end": 1518.16,
    "text": "Okay!"
  },
  {
    "start": 1518.24,
    "end": 1518.74,
    "text": "That is interesting."
  },
  {
    "start": 1518.8,
    "end": 1520.12,
    "text": "you mentioned monocults."
  },
  {
    "start": 1520.18,
    "end": 1522.1,
    "text": "You mentioned chemistry."
  },
  {
    "start": 1522.32,
    "end": 1527.86,
    "text": "AlphaFall was a big topic when you studied or had your first jobs right?"
  },
  {
    "start": 1528.24,
    "end": 1530.82,
    "text": "So there were lot of progress with this topic."
  },
  {
    "start": 1531.2,
    "end": 1538.62,
    "text": "What are so difficult to do in new materials?"
  },
  {
    "start": 1539.08,
    "end": 1545.26,
    "text": "Yeah, so when it comes to using AI for new materials I think you know i mean its an industrial AI podcast right?"
  },
  {
    "start": 1545.44,
    "end": 1546.46,
    "text": "I mean physical AI."
  },
  {
    "start": 1546.62,
    "end": 1549.02,
    "text": "we all know that there is a sim to real gap."
  },
  {
    "start": 1549.42,
    "end": 1558.04,
    "text": "Right When trying to deploy those models in real operating environments and there's some difficulties they're now materials are also physical."
  },
  {
    "start": 1558.1,
    "end": 1568.9,
    "text": "things like We can touch them very much exist In the real world which means we actually suffer the same simtoreal gap that we also see in fields like robotics or other applications."
  },
  {
    "start": 1569.7,
    "end": 1575.68,
    "text": "and this is basically where at Dunya, We've been really focusing on building their type of infrastructure."
  },
  {
    "start": 1575.78,
    "end": 1577.54,
    "text": "We call them autonomous laboratories."
  },
  {
    "start": 1578.36,
    "end": 1582.04,
    "text": "That let you collect these ground-true physical reality data sets."
  },
  {
    "start": 1582.52,
    "end": 1584.46,
    "text": "that basically lets You yeah understand?"
  },
  {
    "start": 1584.52,
    "end": 1590.36,
    "text": "Really In the physical world I how do materials behave How they are being manufactured?"
  },
  {
    "start": 1591.14,
    "end": 1594.26,
    "text": "And yeah, really let you close those gaps."
  },
  {
    "start": 1595.48,
    "end": 1599.32,
    "text": "So the crucial part is again data?"
  },
  {
    "start": 1599.94,
    "end": 1604.84,
    "text": "Yes definitely and then I would argue it's not just about how much data we have."
  },
  {
    "start": 1604.98,
    "end": 1609.16,
    "text": "right It's also what kind of data You know other datasets you generate."
  },
  {
    "start": 1609.2,
    "end": 1610.38,
    "text": "are they relevant to the industry?"
  },
  {
    "start": 1610.54,
    "end": 1619.28,
    "text": "so i think The two big views I see Is for sure seem too real and a second one that may be A little bit more hidden but if anything maybe even bigger."
  },
  {
    "start": 1620.32,
    "end": 1622.76,
    "text": "is the gap between the lab and the FAP, right?"
  },
  {
    "start": 1622.82,
    "end": 1626.7,
    "text": "Like going from a laboratory setting into actual production in industry."
  },
  {
    "start": 1626.86,
    "end": 1628.04,
    "text": "There's still a big gap there."
  },
  {
    "start": 1628.94,
    "end": 1641.82,
    "text": "so it's for sure about like how much data we have generating large volumes of that because you've seen what AI can do when your employed with large datasets such as scale really changed LLMs quite a bit."
  },
  {
    "start": 1642.2,
    "end": 1642.4,
    "text": "now."
  },
  {
    "start": 1642.72,
    "end": 1645.76,
    "text": "For material science I think we foresee something similar."
  },
  {
    "start": 1646.04,
    "end": 1651.74,
    "text": "We're just really early on scaling law And it's really important that we generate or scale the"
  },
  {
    "start": 1651.82,
    "end": 1652.06,
    "text": "right"
  },
  {
    "start": 1652.34,
    "end": 1653.48,
    "text": "type of data as well."
  },
  {
    "start": 1653.52,
    "end": 1656.68,
    "text": "So, not just volume but also I would say quality and relevance."
  },
  {
    "start": 1657.26,
    "end": 1657.44,
    "text": "Okay!"
  },
  {
    "start": 1657.78,
    "end": 1663.42,
    "text": "i want to come back later a bit But can you explain what exactly are your looking at?"
  },
  {
    "start": 1663.68,
    "end": 1665.44,
    "text": "Which industry is focusing on?"
  },
  {
    "start": 1665.8,
    "end": 1671.82,
    "text": "because there guys going into the solar industry Others go in chip manufacturing."
  },
  {
    "start": 1672.18,
    "end": 1675.12,
    "text": "What kind of new materials which industries do they focus?"
  },
  {
    "start": 1677.22,
    "end": 1682.4,
    "text": "One thing first, machine learning is a general purpose technology."
  },
  {
    "start": 1682.72,
    "end": 1689.36,
    "text": "And I think in twenty-twenty five we've started to see the first models emerge that really generalized well"
  },
  {
    "start": 1689.64,
    "end": 1690.06,
    "text": "across"
  },
  {
    "start": 1690.34,
    "end": 1695.4,
    "text": "extremely diverse classes of both molecules and materials right?"
  },
  {
    "start": 1695.44,
    "end": 1703.92,
    "text": "So like there seems to be some kind of representations that these algorithms seem to learn that actually transfers quite well across very large domains at least on the computational end."
  },
  {
    "start": 1703.96,
    "end": 1711.06,
    "text": "That's what industry-relevant applications just yet."
  },
  {
    "start": 1711.72,
    "end": 1722.38,
    "text": "But I think there is some premise that this might actually be a general purpose technology, which was quite exciting when you think about it like really learning what physics does manifest itself in reality."
  },
  {
    "start": 1723.58,
    "end": 1730.56,
    "text": "So that being said, like Dunya is a really small startup company so you know for simple lack of resources You really have to focus."
  },
  {
    "start": 1731.02,
    "end": 1731.84,
    "text": "You can't do everything."
  },
  {
    "start": 1732.12,
    "end": 1737.74,
    "text": "and when we started We looked at a couple things one more mission driven right?"
  },
  {
    "start": 1737.84,
    "end": 1743.88,
    "text": "When we start it Really just look at the biggest problems in the world A lot them being I would say climate related."
  },
  {
    "start": 1744.84,
    "end": 1754.72,
    "text": "Yeah, we believe that novel material innovation could really be a big unlock to potentially massive markets and I think impact right solving biggest problems for society."
  },
  {
    "start": 1754.84,
    "end": 1760.24,
    "text": "We've assumed were gonna make a lot of money doing so And we will maybe little bit even though thinking there."
  },
  {
    "start": 1760.34,
    "end": 1764.3,
    "text": "but let's just like take the hardest part So it can come up with and solve those."
  },
  {
    "start": 1764.52,
    "end": 1764.9,
    "text": "surely that"
  },
  {
    "start": 1764.96,
    "end": 1766.06,
    "text": "was a startup spirit."
  },
  {
    "start": 1767.74,
    "end": 1769.14,
    "text": "So I think this still holds true, right?"
  },
  {
    "start": 1769.3,
    "end": 1771.52,
    "text": "We have one moon shot that we're pursuing internally."
  },
  {
    "start": 1771.84,
    "end": 1774.16,
    "text": "Um...we call it the artificial leaf."
  },
  {
    "start": 1774.46,
    "end": 1777.78,
    "text": "i'm not sure if you've ever read the science fiction novel The Diamond Age."
  },
  {
    "start": 1778.12,
    "end": 1781.64,
    "text": "They have mastered nanotechnology there and they have these printers That can make"
  },
  {
    "start": 1781.82,
    "end": 1783.72,
    "text": "me Yeah yeah yeah"
  },
  {
    "start": 1783.84,
    "end": 1784.88,
    "text": "so were basically working"
  },
  {
    "start": 1785.08,
    "end": 1785.6,
    "text": "on that."
  },
  {
    "start": 1786.6,
    "end": 1788.74,
    "text": "Okay Can you dig a little bit deeper?"
  },
  {
    "start": 1788.82,
    "end": 1789.4,
    "text": "what are you doing?"
  },
  {
    "start": 1789.44,
    "end": 1789.58,
    "text": "Yeah,"
  },
  {
    "start": 1789.72,
    "end": 1791.76,
    "text": "so materials are always part of systems."
  },
  {
    "start": 1791.98,
    "end": 1793.52,
    "text": "So we don't even stop at the material level."
  },
  {
    "start": 1793.64,
    "end": 1797.14,
    "text": "like we do not want to make just powders We really wanna create systems that work."
  },
  {
    "start": 1797.3,
    "end": 1801.28,
    "text": "and of course Material Science plays a huge role in terms of performance What it can do."
  },
  {
    "start": 1802.08,
    "end": 1815.78,
    "text": "But I guess using energy In our case were quite excited about using this fusion reactor in the sky To basically make everything else with right?"
  },
  {
    "start": 1816.08,
    "end": 1818.48,
    "text": "So Molecular, a three-D printer powered by photons."
  },
  {
    "start": 1818.58,
    "end": 1822.06,
    "text": "I think that's maybe how we try to paraphrase it and That's our."
  },
  {
    "start": 1822.4,
    "end": 1825.56,
    "text": "that's our North Star, so this is our single north star in terms of like."
  },
  {
    "start": 1825.6,
    "end": 1833.14,
    "text": "if you're near successful In i don't know Maybe twenty years We'll be able To see such technology actually exist Which Is A bit early yet Right Like We still Live in the Twenty-Twenty."
  },
  {
    "start": 1833.44,
    "end": 1835.76,
    "text": "How Do You Convince The VC To Invest In This?"
  },
  {
    "start": 1836.18,
    "end": 1844.94,
    "text": "Well...You Know..I Think You can make a, I think you could make it good argument that if you were to solve the problem right like yourself."
  },
  {
    "start": 1845.2,
    "end": 1851.8,
    "text": "Two of the biggest problems are mankind and carbon rights or like that itself actually another sort opportunities making network."
  },
  {
    "start": 1852.5,
    "end": 1854.88,
    "text": "And then other thing is when you really dissected."
  },
  {
    "start": 1854.94,
    "end": 1858.34,
    "text": "If your look into some different components they need to be true."
  },
  {
    "start": 1858.84,
    "end": 1865.12,
    "text": "Then we're looking at off-shoot markets that are existing today where this technology could have big impact."
  },
  {
    "start": 1865.26,
    "end": 1874.9,
    "text": "So i guess what convert photons, we need things like electrolytes and membranes to carry charge in transport charge."
  },
  {
    "start": 1875.18,
    "end": 1877.44,
    "text": "So that's relevant for batteries energy technologies."
  },
  {
    "start": 1878.1,
    "end": 1880.58,
    "text": "And then at the core of it you have a lot chemistry."
  },
  {
    "start": 1880.74,
    "end": 1882.0,
    "text": "so you're looking at catalysts."
  },
  {
    "start": 1882.96,
    "end": 1884.68,
    "text": "Catalysts they convert A-to B right?"
  },
  {
    "start": 1885.36,
    "end": 1885.9,
    "text": "They are."
  },
  {
    "start": 1886.24,
    "end": 1888.94,
    "text": "that alone is like a thirty five billion dollar market."
  },
  {
    "start": 1890.9,
    "end": 1898.36,
    "text": "If we are able to develop these tools and apply them into this existing markets today, as we work towards our very ambitious Novsta."
  },
  {
    "start": 1898.62,
    "end": 1904.76,
    "text": "We believe it can create a lot of value for customers while also I guess long term hopefully changing"
  },
  {
    "start": 1904.8,
    "end": 1905.08,
    "text": "the world."
  },
  {
    "start": 1905.56,
    "end": 1910.68,
    "text": "Companies in this field are raising significant amounts of funding."
  },
  {
    "start": 1910.76,
    "end": 1916.94,
    "text": "in the last I would say sixteen months How do you spend your money?"
  },
  {
    "start": 1917.08,
    "end": 1919.0,
    "text": "What is so cost-intensive?"
  },
  {
    "start": 1920.26,
    "end": 1922.16,
    "text": "Yeah, so that's a really interesting question."
  },
  {
    "start": 1922.34,
    "end": 1928.98,
    "text": "I would say to be frank there is for sure a little bit of token-maxing happening in the space"
  },
  {
    "start": 1929.14,
    "end": 1929.62,
    "text": "at the moment?"
  },
  {
    "start": 1930.04,
    "end": 1930.32,
    "text": "Okay!"
  },
  {
    "start": 1930.7,
    "end": 1931.96,
    "text": "Also...I"
  },
  {
    "start": 1932.04,
    "end": 1933.98,
    "text": "think it depends on one."
  },
  {
    "start": 1934.18,
    "end": 1935.52,
    "text": "how do you pose this problem?"
  },
  {
    "start": 1936.16,
    "end": 1940.66,
    "text": "and maybe we might have a different definition from other teams."
  },
  {
    "start": 1941.68,
    "end": 1942.42,
    "text": "What is the difference?"
  },
  {
    "start": 1942.58,
    "end": 1944.52,
    "text": "Was it different to Casper or Antelope?"
  },
  {
    "start": 1944.78,
    "end": 1944.82,
    "text": "So,"
  },
  {
    "start": 1944.9,
    "end": 1947.22,
    "text": "Casper and Topic are great teams in Europe."
  },
  {
    "start": 1947.46,
    "end": 1951.5,
    "text": "I think its definitely worth having a look also what happens on other side of the ocean."
  },
  {
    "start": 1951.56,
    "end": 1951.68,
    "text": "like."
  },
  {
    "start": 1951.72,
    "end": 1954.72,
    "text": "there's great teams like Lila, periodic"
  },
  {
    "start": 1954.8,
    "end": 1955.56,
    "text": "buildings in US."
  },
  {
    "start": 1955.64,
    "end": 1957.24,
    "text": "There're teams from China actually"
  },
  {
    "start": 1958.4,
    "end": 1959.5,
    "text": "being quite advanced as well."
  },
  {
    "start": 1959.56,
    "end": 1959.86,
    "text": "i just"
  },
  {
    "start": 1959.9,
    "end": 1960.18,
    "text": "came back"
  },
  {
    "start": 1960.44,
    "end": 1960.82,
    "text": "over that."
  },
  {
    "start": 1961.46,
    "end": 1962.54,
    "text": "It was very interesting."
  },
  {
    "start": 1963.04,
    "end": 1966.76,
    "text": "But basically...I would say maybe two schools of thought."
  },
  {
    "start": 1968.32,
    "end": 1972.0,
    "text": "Yeah, I think like maybe briefly you have what i would still consider a bit more."
  },
  {
    "start": 1972.14,
    "end": 1974.12,
    "text": "this for me feels a little old right?"
  },
  {
    "start": 1974.16,
    "end": 1979.26,
    "text": "Like this kind of mindset or virtual screening where essentially simulations do a lot of the heavy lifting."
  },
  {
    "start": 1979.92,
    "end": 1988.38,
    "text": "You computationally screen very large amounts of potential materials molecules Maybe tens of millions."
  },
  {
    "start": 1988.44,
    "end": 1995.94,
    "text": "some people are now doing hundreds of trillion and in physical validation part is truly just that."
  },
  {
    "start": 1996.24,
    "end": 1997.3,
    "text": "It's just validation, right?"
  },
  {
    "start": 1997.36,
    "end": 1997.66,
    "text": "So like"
  },
  {
    "start": 1998.8,
    "end": 2000.16,
    "text": "so it's scaling, scaling, scale"
  },
  {
    "start": 2000.28,
    "end": 2011.26,
    "text": "exactly and in the front then you have a very strong argument that given that chemical search space is infinite its really compute bound and compute as what you need to basically scale that."
  },
  {
    "start": 2011.46,
    "end": 2019.9,
    "text": "also if put your X on the basket of doing post-training with frontier LMS this can be really compute intensive AI intensive."
  },
  {
    "start": 2020.16,
    "end": 2022.14,
    "text": "I think both approaches, right?"
  },
  {
    "start": 2022.2,
    "end": 2027.7,
    "text": "Like they basically somewhat assume that a lot of knowledge is already existing today."
  },
  {
    "start": 2028.04,
    "end": 2034.94,
    "text": "That basically help you get to the answer very quickly or in a way that's at least how I see that."
  },
  {
    "start": 2035.1,
    "end": 2038.9,
    "text": "i think this people have been tried before to an extent like it used to be called in silico chemistry."
  },
  {
    "start": 2039.26,
    "end": 2043.34,
    "text": "there was a certain time where that was really fashionable to try to use computational methods alone."
  },
  {
    "start": 2043.62,
    "end": 2045.48,
    "text": "that hasn't had so much benefits."
  },
  {
    "start": 2045.56,
    "end": 2051.86,
    "text": "and now we have a renaissance using machine learning which gives us superpowers much more ambitious here."
  },
  {
    "start": 2052.08,
    "end": 2055.9,
    "text": "But then again, we have the same kind of issues around the sim to real gap."
  },
  {
    "start": 2055.98,
    "end": 2063.219,
    "text": "this lab too fast right like actually getting things to work in production is a very different beast and I think The Other School Of Thought Is A Bit More"
  },
  {
    "start": 2063.44,
    "end": 2063.6,
    "text": "This."
  },
  {
    "start": 2063.719,
    "end": 2065.04,
    "text": "That's The Berlin Approach."
  },
  {
    "start": 2065.1,
    "end": 2065.239,
    "text": "Now"
  },
  {
    "start": 2065.6,
    "end": 2066.699,
    "text": "Let'S Call It The Berlin approach."
  },
  {
    "start": 2066.78,
    "end": 2067.6,
    "text": "i Like That Yeah!"
  },
  {
    "start": 2068.199,
    "end": 2080.34,
    "text": "The Berlin School Of thought Is Very Much The Idea Of More Of An Active Learning Or Self Reinforcing Approach where indeed, every experiment compounds."
  },
  {
    "start": 2081.08,
    "end": 2087.699,
    "text": "Every time you go to the laboratory and generate new data that helps make better decisions in future."
  },
  {
    "start": 2088.0,
    "end": 2099.48,
    "text": "That I think essentially addresses some of these simtoreal and lab-to-fab constraints explicitly And i think this is an exponential technology really which will only get better."
  },
  {
    "start": 2100.2,
    "end": 2107.66,
    "text": "This approach will most likely succeed And then I think it depends a little bit on that spectrum."
  },
  {
    "start": 2117.12,
    "end": 2131.8,
    "text": "And I think scale is almost always the answer in startup land, but like knowing what to scale it's really tough."
  },
  {
    "start": 2132.04,
    "end": 2137.22,
    "text": "Exactly and want to come back to that data generation can you walk us through the process?"
  },
  {
    "start": 2137.7,
    "end": 2139.0,
    "text": "What makes use so special?"
  },
  {
    "start": 2139.46,
    "end": 2148.38,
    "text": "Yeah So i think there was a rather nascent technology called autonomous labs or self-driving labs."
  },
  {
    "start": 2149.2,
    "end": 2153.94,
    "text": "It used be a little bit of confusing metaphor because those labs don't necessarily drive around."
  },
  {
    "start": 2154.68,
    "end": 2156.82,
    "text": "It's like the robotics gyms, or what is it?"
  },
  {
    "start": 2157.08,
    "end": 2158.92,
    "text": "So we see that right here."
  },
  {
    "start": 2158.96,
    "end": 2159.72,
    "text": "That and the robotics."
  },
  {
    "start": 2159.78,
    "end": 2159.98,
    "text": "yet"
  },
  {
    "start": 2160.12,
    "end": 2166.4,
    "text": "in a way yeah imagine basically facility right physical facility which has a room full of robots."
  },
  {
    "start": 2166.74,
    "end": 2181.0,
    "text": "those robots are connected to scientific instruments And they can basically do what PhD chemist or material scientist would be doing manually In one high throughput fashion but two also in a very software-first programmable way."
  },
  {
    "start": 2181.4,
    "end": 2190.02,
    "text": "Not all autonomous laps or self driving laps are created equal, that being said it's been technology that has matured over the last ten years."
  },
  {
    "start": 2190.34,
    "end": 2194.14,
    "text": "There have been a lot of really exciting academic breakthroughs on that topic."
  },
  {
    "start": 2194.22,
    "end": 2204.3,
    "text": "So many, many exciting demonstrations about technology can do for relatively speaking simple problems and increasingly so now we're industrializing that I would say yes probably at the very forefront off that"
  },
  {
    "start": 2204.38,
    "end": 2204.68,
    "text": "movement."
  },
  {
    "start": 2205.04,
    "end": 2205.54,
    "text": "interrupt you."
  },
  {
    "start": 2205.6,
    "end": 2207.88,
    "text": "Can You walk us through the process how it starts?"
  },
  {
    "start": 2208.38,
    "end": 2209.5,
    "text": "And what are your doing then?"
  },
  {
    "start": 2209.68,
    "end": 2209.8,
    "text": "yeah"
  },
  {
    "start": 2210.36,
    "end": 2213.06,
    "text": "with this autonomous apps essentially What'd he come?"
  },
  {
    "start": 2213.3,
    "end": 2213.9,
    "text": "How could imagine?"
  },
  {
    "start": 2214.06,
    "end": 2219.46,
    "text": "right like y'know they did as two things try to invent new material But It also generates data in The process Of Doing."
  },
  {
    "start": 2220.22,
    "end": 2227.1,
    "text": "So for each data point, essentially what you get is a material which has sample history."
  },
  {
    "start": 2227.82,
    "end": 2229.98,
    "text": "How was that material manufactured?"
  },
  {
    "start": 2230.18,
    "end": 2234.96,
    "text": "What were all the different processing steps that went into it like pre-curses and how they processed them?"
  },
  {
    "start": 2235.84,
    "end": 2236.32,
    "text": "The temperature?"
  },
  {
    "start": 2236.38,
    "end": 2241.88,
    "text": "maybe it was annealed at across those processes where we can generate additional characterization data."
  },
  {
    "start": 2242.3,
    "end": 2245.7,
    "text": "so photograph of it, you do some spectroscopy."
  },
  {
    "start": 2246.14,
    "end": 2262.88,
    "text": "You look at measurements like what's the conductivity and you get performance data out of it, right?"
  },
  {
    "start": 2262.92,
    "end": 2265.14,
    "text": "So your looking at performance whatever that may mean."
  },
  {
    "start": 2265.94,
    "end": 2270.22,
    "text": "Maybe catalytic properties maybe some solar efficiency for the solar cell."
  },
  {
    "start": 2270.52,
    "end": 2272.74,
    "text": "And then you also look things like stability Right."
  },
  {
    "start": 2272.8,
    "end": 2278.42,
    "text": "so real degradation For example is The top one question we get when We speak to industrial partners."
  },
  {
    "start": 2279.18,
    "end": 2284.52,
    "text": "And ultimately That's something a lot Of machine learning models Out there today are not very good at predicting It."
  },
  {
    "start": 2285.26,
    "end": 2287.66,
    "text": "Some properties simply empirical."
  },
  {
    "start": 2287.84,
    "end": 2296.2,
    "text": "No professor in the world, no LLM can tell you if you take five metal precursors and make an alloy out of them what's a conductivity?"
  },
  {
    "start": 2297.3,
    "end": 2297.84,
    "text": "That's very tough."
  },
  {
    "start": 2298.56,
    "end": 2301.52,
    "text": "so that's basically the type of data that the lab produces."
  },
  {
    "start": 2302.16,
    "end": 2305.4,
    "text": "we cannot augment that with simulations right simulations?"
  },
  {
    "start": 2305.44,
    "end": 2309.48,
    "text": "In our context they often start with first principle physics simulations."
  },
  {
    "start": 2309.86,
    "end": 2312.2,
    "text": "So it is actually approximating the Schrodinger equation."
  },
  {
    "start": 2313.22,
    "end": 2325.38,
    "text": "You can use different levels of theory and like you know, use multi-scale physics ultimately to learn rich representations about those materials are could be so materials that kind of funky as in like humans have a notion on what the material is."
  },
  {
    "start": 2325.86,
    "end": 2327.08,
    "text": "It's actually quite hard to describe."
  },
  {
    "start": 2327.14,
    "end": 2332.9,
    "text": "So let your only ever observe certain behaviors after material with the true nature of it."
  },
  {
    "start": 2333.54,
    "end": 2334.96,
    "text": "That's how we frame the problem."
  },
  {
    "start": 2335.12,
    "end": 2363.04,
    "text": "And just for contrast I think when the machine learning community started to get excited about it, But in the real world, that's really complicated because we don't have to start replicating a machine."
  },
  {
    "start": 2363.08,
    "end": 2364.72,
    "text": "That can just like place atoms away."
  },
  {
    "start": 2364.78,
    "end": 2365.16,
    "text": "We want right?"
  },
  {
    "start": 2365.26,
    "end": 2368.1,
    "text": "Like we don' t have them for industry and it doesn''t work."
  },
  {
    "start": 2368.74,
    "end": 2376.02,
    "text": "So this is what I mean when I say Dunja or The Bullen School of Thought poses the problem A little bit differently Is basically about that."
  },
  {
    "start": 2376.12,
    "end": 2378.76,
    "text": "so we look at materials i guess In study different way."
  },
  {
    "start": 2378.94,
    "end": 2389.04,
    "text": "use simulation Really help us understand better What these real-world Materials actually are how they behave And then build a knowledge graph, right?"
  },
  {
    "start": 2389.1,
    "end": 2392.24,
    "text": "Like when the knowledge graph we aggregate our experimental results."
  },
  {
    "start": 2392.3,
    "end": 2398.2,
    "text": "We aggregate prior knowledge from let's say pattern databases and also than this emulation results."
  },
  {
    "start": 2398.52,
    "end": 2412.18,
    "text": "on top of that we can do two things like we can basically train domain specific models or scaling up sort of data volume which generating so were going towards foundation model scale Which is really exciting."
  },
  {
    "start": 2412.22,
    "end": 2417.02,
    "text": "And then the other thing is you can literally use agents and you can use any type of agent."
  },
  {
    "start": 2417.1,
    "end": 2422.34,
    "text": "So we've built our own, We have also used Claude, various other models etc."
  },
  {
    "start": 2423.24,
    "end": 2428.5,
    "text": "to basically look at that data and come up with a new scientific hypothesis."
  },
  {
    "start": 2428.72,
    "end": 2429.6,
    "text": "Does it really work?"
  },
  {
    "start": 2429.7,
    "end": 2436.0,
    "text": "Yeah I mean the first agentic discoveries of new materials."
  },
  {
    "start": 2436.16,
    "end": 2437.76,
    "text": "I think that is actually really, really exciting."
  },
  {
    "start": 2437.82,
    "end": 2441.52,
    "text": "so i think this from my perspective very much frontier yeah"
  },
  {
    "start": 2441.94,
    "end": 2443.16,
    "text": "sounds like"
  },
  {
    "start": 2443.22,
    "end": 2444.26,
    "text": "a lot of people want to do that."
  },
  {
    "start": 2444.32,
    "end": 2447.72,
    "text": "but yeah i think dunia now again because we made the choices we've made in the past."
  },
  {
    "start": 2448.34,
    "end": 2451.44,
    "text": "uh We are at a position where we really build an API for the physical world."
  },
  {
    "start": 2451.58,
    "end": 2464.08,
    "text": "So you can give an AI scientist and environment Where it starts to explore come up with ideas Validate those ideas get the data back within twenty four hours and really run campaigns that way."
  },
  {
    "start": 2464.14,
    "end": 2465.12,
    "text": "And like, yeah we use that."
  },
  {
    "start": 2465.24,
    "end": 2467.3,
    "text": "it generated a bunch of different hypotheses."
  },
  {
    "start": 2467.38,
    "end": 2477.08,
    "text": "some of them It could confirm Some of them didn't write but I guess this is sort of very much scientific method in action and using That We actually found some patent worthy Catalyst materials."
  },
  {
    "start": 2477.14,
    "end": 2477.9,
    "text": "In that case right."
  },
  {
    "start": 2477.96,
    "end": 2479.76,
    "text": "so we found some catalysts."
  },
  {
    "start": 2479.8,
    "end": 2479.9,
    "text": "well?"
  },
  {
    "start": 2479.94,
    "end": 2480.46,
    "text": "That's interesting."
  },
  {
    "start": 2480.74,
    "end": 2484.14,
    "text": "can you describe at The end A little bit what?"
  },
  {
    "start": 2484.2,
    "end": 2487.04,
    "text": "What does your business model look like?"
  },
  {
    "start": 2487.4,
    "end": 2496.16,
    "text": "Do you work closely with companies or do you want to have a license-based stuff, is it a search engine?"
  },
  {
    "start": 2497.46,
    "end": 2499.52,
    "text": "and what's the business model."
  },
  {
    "start": 2505.12,
    "end": 2511.06,
    "text": "essentially in AI and like combining it with these physical systems that actually do this data set to let you solve those problems."
  },
  {
    "start": 2511.1,
    "end": 2512.82,
    "text": "And then also find new materials, right?"
  },
  {
    "start": 2513.0,
    "end": 2521.86,
    "text": "So what we bring to the table things that were not good at is large scale manufacturing or ultimately... We of course look at end customers."
  },
  {
    "start": 2522.04,
    "end": 2524.74,
    "text": "they're really excited by these material science use cases."
  },
  {
    "start": 2525.78,
    "end": 2530.38,
    "text": "so yeah looking for chemical industry partners on what's the main focus"
  },
  {
    "start": 2530.8,
    "end": 2542.1,
    "text": "Yeah, so I think we identified a few segments but for sure the chemical industry understands what we're doing and they are quite excited about like uh What could be true if you truly able to solve some of those challenges right?"
  },
  {
    "start": 2542.14,
    "end": 2547.06,
    "text": "So they understand very well this is very hard But then also understand that there need new products."
  },
  {
    "start": 2547.14,
    "end": 2552.02,
    "text": "New high-margined products, new markets And i think traditional innovation cycles take very long."
  },
  {
    "start": 2552.78,
    "end": 2555.24,
    "text": "So anything that can improve productivity Is already great!"
  },
  {
    "start": 2555.62,
    "end": 2560.72,
    "text": "If your truly able find new materials That can have massive as well."
  },
  {
    "start": 2560.92,
    "end": 2570.14,
    "text": "So they understand that there is a company said, maybe not necessarily considered chemical companies but like companies are very much depend on material performance to make their products work right?"
  },
  {
    "start": 2570.18,
    "end": 2577.76,
    "text": "So I think semiconductor industry or like battery industry or other type of industries where material performances just take very much at the core for what?"
  },
  {
    "start": 2578.82,
    "end": 2581.94,
    "text": "And you're looking for partnerships there, or do want to be a tech partner?"
  },
  {
    "start": 2582.52,
    "end": 2583.18,
    "text": "Exactly."
  },
  {
    "start": 2583.72,
    "end": 2588.34,
    "text": "So how we think about the business model is very much driven by partnering with industry."
  },
  {
    "start": 2595.46,
    "end": 2596.94,
    "text": "The problems they see the pain points."
  },
  {
    "start": 2596.98,
    "end": 2608.98,
    "text": "They have when it comes to their performance bottlenecks, but in terms of you know also I guess at this point and time we Have a lot of supply chain disruption there's A lot of other requests were getting is like hey can We reformulate this product without This component?"
  },
  {
    "start": 2609.06,
    "end": 2610.14,
    "text": "or like change metal?"
  },
  {
    "start": 2610.3,
    "end": 2613.52,
    "text": "And i don't have people use or You Know do something about."
  },
  {
    "start": 2614.0,
    "end": 2616.74,
    "text": "yeah supply chain Like issues are regulation issues."
  },
  {
    "start": 2617.12,
    "end": 2622.16,
    "text": "so those types Of request that usually would take you know many years maybe decade-long Research programs."
  },
  {
    "start": 2622.2,
    "end": 2624.2,
    "text": "then the question simply Is he can we do this faster?"
  },
  {
    "start": 2625.42,
    "end": 2629.36,
    "text": "your technology to basically run such a campaign at accelerated speed."
  },
  {
    "start": 2629.66,
    "end": 2638.72,
    "text": "So it's like very co-development driven, the one thing that we discovered on the business side other than the science is also this infrastructure we built turns out seems."
  },
  {
    "start": 2638.76,
    "end": 2644.48,
    "text": "So when we started again, like we just wanted to invent new materials and sell that."
  },
  {
    "start": 2644.6,
    "end": 2648.42,
    "text": "Like there was very much the startup mindset in the beginning right?"
  },
  {
    "start": 2648.48,
    "end": 2648.98,
    "text": "It's okay."
  },
  {
    "start": 2649.26,
    "end": 2651.98,
    "text": "When we did this discovery company were going to just sell material sets."
  },
  {
    "start": 2652.08,
    "end": 2653.62,
    "text": "That is a thing you can come up with."
  },
  {
    "start": 2653.82,
    "end": 2657.42,
    "text": "what We learned now Is that these other elements are described."
  },
  {
    "start": 2657.46,
    "end": 2661.32,
    "text": "try these databases as proprietary data sets The lab infrastructure."
  },
  {
    "start": 2661.36,
    "end": 2663.46,
    "text": "that itself actually has quite A bit of value."
  },
  {
    "start": 2663.6,
    "end": 2673.14,
    "text": "This something that um You know our partners seem To really appreciate And there is a whole new segment of companies now that works on material science, so people may not quite realize."
  },
  {
    "start": 2673.48,
    "end": 2673.64,
    "text": "Right?"
  },
  {
    "start": 2673.86,
    "end": 2678.02,
    "text": "So my fans right Tiktok are doing battery materials."
  },
  {
    "start": 2678.12,
    "end": 2686.44,
    "text": "now Amazon Alice's DeepMind is building a lab in London that does material science."
  },
  {
    "start": 2686.52,
    "end": 2689.66,
    "text": "Thropic, of course has big effort on AI for Science."
  },
  {
    "start": 2689.72,
    "end": 2691.24,
    "text": "like there are just lot of cloud signs."
  },
  {
    "start": 2691.74,
    "end": 2693.28,
    "text": "they're now developing drugs right?"
  },
  {
    "start": 2693.36,
    "end": 2695.08,
    "text": "A lot people think their materials are next."
  },
  {
    "start": 2695.38,
    "end": 2706.22,
    "text": "so all these big tech companies that essentially moving into the physical world and I think Materials is one of this trillion dollar frontiers where People thinks it could be really exciting application of AI."
  },
  {
    "start": 2706.7,
    "end": 2714.44,
    "text": "being said as you mentioned earlier its'all about data And one of the very few players that knows how to generate reasonably large data sets."
  },
  {
    "start": 2714.94,
    "end": 2718.88,
    "text": "That can be used for really learning, how materials again are manufactured right?"
  },
  {
    "start": 2719.08,
    "end": 2720.9,
    "text": "How it's synthesized or so...how do they behave like?"
  },
  {
    "start": 2720.94,
    "end": 2723.8,
    "text": "How did they degrade and actually behave in operating systems?"
  },
  {
    "start": 2723.88,
    "end": 2725.5,
    "text": "maybe you're becoming a data provider."
  },
  {
    "start": 2725.58,
    "end": 2726.68,
    "text": "It is something we've been exploring."
  },
  {
    "start": 2726.74,
    "end": 2733.18,
    "text": "So We had already first paying customer using a model which was closer to that mindset."
  },
  {
    "start": 2733.56,
    "end": 2735.76,
    "text": "I think this is something nobody really know."
  },
  {
    "start": 2735.9,
    "end": 2738.32,
    "text": "so i think donia continues to operate on their mind set."
  },
  {
    "start": 2738.7,
    "end": 2741.96,
    "text": "Okay, let's just solve big problems for the community and that will pay off."
  },
  {
    "start": 2742.36,
    "end": 2749.24,
    "text": "And I think we see those two vectors where and it's new material innovations."
  },
  {
    "start": 2749.4,
    "end": 2751.76,
    "text": "What is on your agenda in the next coming months, Alexander?"
  },
  {
    "start": 2751.82,
    "end": 2751.96,
    "text": "Next"
  },
  {
    "start": 2752.02,
    "end": 2752.74,
    "text": "coming month now."
  },
  {
    "start": 2752.92,
    "end": 2758.74,
    "text": "so I actually am going on a hiring spree... So you will be listening and once-a-job."
  },
  {
    "start": 2758.96,
    "end": 2759.86,
    "text": "please reach out!"
  },
  {
    "start": 2761.16,
    "end": 2774.28,
    "text": "We'll be hiring a lot and probably its good podcast to be on because as i mentioned these physical facilities of course we're hiring people with AI backgrounds but within Dunia really need manufacturing engineers."
  },
  {
    "start": 2774.34,
    "end": 2781.8,
    "text": "systems engineers know about factory automation, that know about I guess you know operational excellence when it comes to those types of facilities."
  },
  {
    "start": 2782.34,
    "end": 2785.42,
    "text": "because again we're essentially generating these data sets."
  },
  {
    "start": 2785.46,
    "end": 2800.86,
    "text": "We are building an essentially a factory that does science and its very different from traditional manufacturing where the factory usually makes never quite makes the same product twice, you know?"
  },
  {
    "start": 2801.02,
    "end": 2803.1,
    "text": "I mean apart from replication and reproducibility."
  },
  {
    "start": 2803.34,
    "end": 2806.04,
    "text": "So that's the industry of four point zero dream right."
  },
  {
    "start": 2806.16,
    "end": 2807.56,
    "text": "so individual production"
  },
  {
    "start": 2807.72,
    "end": 2808.42,
    "text": "Yeah essentially!"
  },
  {
    "start": 2808.66,
    "end": 2809.84,
    "text": "That is like what we are talking about."
  },
  {
    "start": 2809.88,
    "end": 2822.64,
    "text": "We're doing it on a lab scale to be flexible To basically try many different types of chemistries for given products And i think from an engineering perspective there is lot very interesting challenges."
  },
  {
    "start": 2822.8,
    "end": 2822.92,
    "text": "come"
  },
  {
    "start": 2822.96,
    "end": 2824.68,
    "text": "with them Alexander, it was a pleasure."
  },
  {
    "start": 2824.76,
    "end": 2826.38,
    "text": "Thank you very much and all the best for you!"
  },
  {
    "start": 2826.46,
    "end": 2827.6,
    "text": "We keep our fingers crossed."
  },
  {
    "start": 2828.14,
    "end": 2831.78,
    "text": "Very inspiring...very impressive ideas & concepts."
  },
  {
    "start": 2831.98,
    "end": 2833.22,
    "text": "Thanks alot for recording"
  },
  {
    "start": 2833.68,
    "end": 2833.92,
    "text": "Awesome."
  },
  {
    "start": 2834.06,
    "end": 2834.68,
    "text": "thankyou so much."
  },
  {
    "start": 2834.72,
    "end": 2838.06,
    "text": "Have A Good One Bye."
  }
]