Latest / Code & Cure / #67 - How AI Helps Make mRNA Vaccines Thermostable

Transcript

#67 - How AI Helps Make mRNA Vaccines Thermostable · Code & Cure

0:00
SPEAKER_01: The next breakthrough in mRNA vaccines might not be what they protect you against, but how well they withstand room temperature.
0:09
SPEAKER_01: Can AI help us get there?
0:20
SPEAKER_00: Hello and welcome back to Coding Cure, where we discuss decoding health in the age of AI.
0:25
SPEAKER_00: My name is Vasant Sarati.
0:27
SPEAKER_00: I'm an AI researcher and cognitive scientist, and I'm here as always with Laura Hagopian.
0:31
SPEAKER_01: I'm an emergency medicine physician and I work in digital health.
0:35
SPEAKER_00: So I'm very, very excited about this paper.
0:38
SPEAKER_00: I know I say that a lot, but I'm You're excited about every paper.
0:40
SPEAKER_01: You're like, there's a paper?
0:42
SPEAKER_01: I'm excited.
0:43
SPEAKER_00: Well, the yeah, this one I'm excited because it's sort of uh a little like a fresh break from a lot of the discussions we've had recently about LLMs and safety and security.
0:52
SPEAKER_00: Doom and gloom.
0:53
SPEAKER_00: A lot of doom and gloom discussions and such.
0:54
SPEAKER_00: And this is just a genuinely clean, interesting paper with an interesting approach and providing something that's super useful.
1:02
SPEAKER_01: It actually feels like a breakthrough and it feels like um it has the potential to impact so many people at the population level in a really positive way from like the public health standpoint.
1:13
SPEAKER_00: Yes, exactly.
1:13
SPEAKER_00: And I just want to alert the re-uh the listeners that there is no LLM in this approach.
1:19
SPEAKER_00: Uh it is called AI, but there is no LLM.
1:21
SPEAKER_00: There's no, you know, worry about chatbots or anything else.
1:25
SPEAKER_01: I mean, sometimes like the best ideas are some of the older ones.
1:29
SPEAKER_01: Like I think people use artificial intelligence and large language models interchangeably, but a lot of the the older stuff that's out there, I know we're gonna be talking about Bayesian algorithms today, is like, hey, it's really cool, it's really good.
1:44
SPEAKER_01: It fits this use case in particular very well.
1:47
SPEAKER_01: And um, and I I feel like we're just like skirting around what happened in this paper, but yeah, let's get right to it.
1:53
SPEAKER_01: But let's get to it.
1:54
SPEAKER_01: Yeah.
1:55
SPEAKER_01: What they what they did in this paper was they figured out a way to make mRNA vaccines thermostable.
2:03
SPEAKER_00: What does thermostable mean?
2:04
SPEAKER_01: Stable at room temperature.
2:06
SPEAKER_00: Stable at room temperature.
2:07
SPEAKER_00: Okay.
2:07
SPEAKER_01: So the whole so so basically, like if you have, I don't know if you remember back to the days of COVID.
2:13
SPEAKER_01: Uh I know COVID's still around, but when the vaccines were first being developed with mRNA, one of the biggest problems with distributing the vaccines was these, were these cold chain requirements, where the vaccine had to be stored at like very cold temperatures, which means that like it has to be transported at those very cold temperatures.
2:36
SPEAKER_01: When it gets to the spot it's going to, whether that's like a clinic or a pharmacy, it has to be stored at those temperatures.
2:43
SPEAKER_01: You don't want to have it stored at a higher temperature, or it could go bad.
2:47
SPEAKER_01: And so it needs like all of this special packaging.
2:50
SPEAKER_01: It needs like transportation conditions that are amenable to that.
2:54
SPEAKER_01: And you have to monitor it the whole time to make sure it has stayed at those temperatures.
2:59
SPEAKER_01: And so that's like a big deal because not everywhere has that, right?
3:04
SPEAKER_01: Like there are some countries where like you're like, well, I don't even think we could do this.
3:09
SPEAKER_01: And places where they do have that, like we do have this in the United States, it's expensive.
3:14
SPEAKER_01: It's like a big deal.
3:15
SPEAKER_01: And so the COVID vaccine is a great example of a of an mRNA vaccine where if you store it at like a little bit above freezing, it can last for two months.
3:29
SPEAKER_01: And if you store it at a very cold temperature, like negative 50 to negative 15 degrees Celsius, it can last for 10 months.
3:36
SPEAKER_01: But that's the storage temperatures that you're looking at.
3:39
SPEAKER_01: And what happens is when there is this ultra cold temperature requirement, it's really hard to get the vaccine into limited resource settings.
3:52
SPEAKER_01: And there's a lot of waste that happens because things happen along the way where the storage may not be perfect.
3:58
SPEAKER_01: And then you lose the vaccine.
3:59
SPEAKER_01: And now you lose or the vaccine expires.
4:01
SPEAKER_01: Exactly.
4:02
SPEAKER_01: Right.
4:02
SPEAKER_01: And so the thought has always been hey, can we make an mRNA vaccine that is stable at room temperature?
4:13
SPEAKER_01: Because this would make it not only safe, but like scalable.
4:17
SPEAKER_01: Now we can easily transport it.
4:19
SPEAKER_01: It doesn't, it's not so expensive to do so, but we can get it to places that don't have these ultra-cold storage, you know, special coolers or whatever it is, right?
4:30
SPEAKER_01: Right.
4:31
SPEAKER_01: And we want it to be stable at that room temperature for, you know, a period of time.
4:37
SPEAKER_01: So we can transport it and get it there.
4:39
SPEAKER_01: And that way we can have a longer shelf life and it would it would be better, right?
4:47
SPEAKER_00: Right, right, right, right, right.
4:48
SPEAKER_00: No, exactly.
4:49
SPEAKER_00: And so that's the space in which they're working.
4:51
SPEAKER_00: And, you know, I think that one of the challenges is that you have to do a lot of experimentation to figure out how to make it last in room temperatures for a long time.
5:01
SPEAKER_01: Right.
5:01
SPEAKER_01: Like people haven't really figured this out yet.
5:03
SPEAKER_01: There have been ideas, they've tried different things, um, but they don't have an answer.
5:09
SPEAKER_01: And so so the main reason to do this is that you're gonna make it easier to store, you're gonna make it easier to transport.
5:16
SPEAKER_01: But there's also like a secondary reason that goes along with this, which is that um based on the way that they are doing it in this paper, you could actually deliver the vaccine without a needle.
5:31
SPEAKER_00: Oh, interesting.
5:32
SPEAKER_00: Yeah.
5:32
SPEAKER_00: Oh, yes.
5:33
SPEAKER_00: That's that's the part about it being solid state, right?
5:36
SPEAKER_00: So we're versus being a liquid form or whatever.
5:39
SPEAKER_01: Yeah.
5:40
SPEAKER_01: There the solid state version of this can be integrated into these little patches that you can just like put on someone, which is cool on many levels.
5:50
SPEAKER_01: One is like some people do not like needles, so they don't want to get a vaccine with a needle, right?
5:55
SPEAKER_01: But two is when you give a vaccine with a needle, you need someone who knows how to give a vaccine with a needle.
6:02
SPEAKER_01: If you do it with patches, well, now you've expanded vaccine access because you don't need to have a sharp waste container and you don't need to have people who are trained in how to give a vaccine with a needle.
6:14
SPEAKER_01: Got it.
6:15
SPEAKER_01: And so this is really cool from multiple angles because we could, in theory, get the vaccines to more people more easily because they're stable at room temperature and easier to transport, and potentially deliver them in an easier fashion that doesn't require trained personnel or sharp swast and get them to people who are scared to get vaccines because they don't like needles.
6:39
SPEAKER_01: Right, right.
6:41
SPEAKER_01: So I think from like a public health perspective, globally, this is like a really cool concept.
6:48
SPEAKER_01: This is what we want.
6:50
SPEAKER_01: And I'm interested to hear from you, Vasanth, like, how do they use AI to do this?
6:55
SPEAKER_01: Because one of the things that we talk a lot about with AI is like, oh, you need to give it lots of data and it's gonna figure out an answer.
7:03
SPEAKER_01: But in this case, there's actually like not a big, big data set to feed it.
7:08
SPEAKER_01: Like we haven't really done this before.
7:10
SPEAKER_01: Yeah, it's not been successful.
7:11
SPEAKER_01: They've tried some things that didn't work very well.
7:14
SPEAKER_01: So it's like, well, how do you go about thinking using AI to solve a problem like this?
7:19
SPEAKER_00: Right.
7:19
SPEAKER_00: And so I think that's a that's that's it.
7:21
SPEAKER_00: I think that's at the core of the question.
7:23
SPEAKER_00: And I think before we dig into the AI, there's one little tiny piece of biology I think that we have to we have to hit on, which is that the notion of uh excipients, excuse me, uh, that are uh chemicals that are added um or additives that are added to the formulation of the vaccine that isn't part of the drug.
7:43
SPEAKER_00: And um, you know, I I think the example would be things like sugars and salts and polymers and other things um that basically allow you to protect the active ingredient that's in there.
7:53
SPEAKER_00: Um so that's a very cool, important point because the uh paper itself talks about uh four or five different key of key uh varieties of these things, including things like glucose and sucrose and some other things here.
8:08
SPEAKER_00: Um but why I'm talking about these uh incipients um is that that's where the design is.
8:16
SPEAKER_00: People have to figure out that the point of the uh getting the uh vaccine to be the right consistency and the and handle be thermostable and all that stuff depends on the right formulation of these other chemicals.
8:28
SPEAKER_00: And how much you put off those other um, you know, uh compounds is what drives the entire design of this.
8:36
SPEAKER_01: So then in theory, like what we're not changing like the mRNA, you know, the thing that is actually you're vaccinating with, you're changing like what's around it.
8:45
SPEAKER_00: Yes.
8:46
SPEAKER_00: You know, I I I in my head, I don't know if this is the right analogy, but in my head, I think of this as being like when you pickle something, right?
8:52
SPEAKER_00: You're trying to keep it for longer, it's at the very core of it, it's a piece of cucumber that you put in, but then you add all these other things to make it last longer.
9:00
SPEAKER_01: Okay, but like in theory, then also you can pickle like celery and carrots and whatever too.
9:05
SPEAKER_01: So in theory, if like we get a good pickling solution, then it should work in other settings too.
9:11
SPEAKER_00: We should be able to pickle other mRNA vaccines.
9:14
SPEAKER_00: No, no, absolutely.
9:15
SPEAKER_00: That's I mean, that's part of the point, right?
9:16
SPEAKER_00: So you have these, I mean, outside of this, I don't even know if this pickle analogy actually works, but right now it does.
9:22
SPEAKER_00: So let's just go with it.
9:24
SPEAKER_00: But uh, but the core idea is you have all these other things, these uh ingredients.
9:28
SPEAKER_01: It doesn't, you know why it doesn't quite work?
9:29
SPEAKER_01: Because we're talking about getting things into a solid state formulation.
9:32
SPEAKER_00: Oh, I suppose it doesn't really work.
9:34
SPEAKER_00: Yeah.
9:34
SPEAKER_01: And like I'm just thinking about the pick pickle juice.
9:36
SPEAKER_01: Oh boy.
9:37
SPEAKER_00: Yeah.
9:38
SPEAKER_00: Okay, let's let's forget the pickle analogy for a little bit.
9:40
SPEAKER_00: It's I think it's done its job.
9:42
SPEAKER_00: Um, so we have all these chemicals that we want to add, and and I call them chemicals, but it's just these additives that you want to add to the mRNA um core that makes the formulation have certain properties.
9:53
SPEAKER_00: And but you don't know how much to add to these things.
9:55
SPEAKER_00: And that's the design space.
9:57
SPEAKER_00: And what I mean by design space is that you want to decide how much, what you know, like levels you want to add each of these, and different levels perform differently for different purposes.
10:06
SPEAKER_00: And so it's almost like you have all of these knobs that you have to turn to kind of let in the right amount of these different additives uh to make the right recipe.
10:15
SPEAKER_01: So you're like trying to figure out you're like, oh, I have these five additives, like what's the correct ratio to get this in the solid state?
10:21
SPEAKER_00: Yeah.
10:21
SPEAKER_00: So basically what you do is you you you you you you have this um machine, let's say you have this pretend pretend machine where you throw in all the ingredients, if you dry it, bake it, and um you bake it at 37 degrees Celsius for weeks, and then you measure how the drug still works or not, right?
10:38
SPEAKER_01: Because that's that's typically the You're just like going on with the food analogies, yeah.
10:42
SPEAKER_00: Maybe I'm hungry, I don't know.
10:44
SPEAKER_00: Um But but the problem is if if say you came up with some set of ratios of these additives and you put it into this box and you made the uh perfect mixture, you think it's right, it's going to take you weeks before you know that it even worked.
11:00
SPEAKER_00: And then after 37 degrees for several weeks, you realize you get the thing back and it's you have to physically run the thing, and then you figure out that maybe it doesn't work, and then you have to change it again.
11:10
SPEAKER_01: So what you're telling me is it takes you like two months to make cupcakes.
11:15
SPEAKER_00: Yes.
11:16
SPEAKER_00: Which which is about the above.
11:17
SPEAKER_01: I think you need to go back to the pickle analogy because that actually takes a long time.
11:22
SPEAKER_00: It does.
11:22
SPEAKER_00: And so the point is that these evaluations of your design choices are expensive uh because uh each evaluation costs time and money.
11:33
SPEAKER_00: I mean, sorry, uh yeah, time and money.
11:35
SPEAKER_00: And so you have you have to wait a while for it.
11:38
SPEAKER_00: Um and it's not like your traditional machine learning problem where you get a result and you can sort of feed it back through the neural network and update its uh uh values and then do it again really fast.
11:50
SPEAKER_00: You can't do that, you can't do multiple iterations so much.
11:53
SPEAKER_01: You have so the pro so then you want to get it right from the beginning, right?
11:56
SPEAKER_00: You want to either get it right from the beginning or do experimentation fast in a smart way that gets you to the answer as fast as you can possibly get to the answer, right?
12:06
SPEAKER_00: Uh and so you you can't just like blindly search all the possibilities.
12:11
SPEAKER_01: Okay, so how do they sprinkle magic powder on this?
12:13
SPEAKER_00: Right.
12:14
SPEAKER_00: And I mean, that is the question, right?
12:15
SPEAKER_00: Um, you you also don't necessarily, I mean, although you want the best recipe, you don't know what that is upfront.
12:23
SPEAKER_01: This is where I'm this is where the question mark is for me, because a lot of times we talk about inputting tons of data, and here there's like you don't have tons of data to input, and what they've done so far hasn't worked.
12:34
SPEAKER_01: So how is it like magically like figuring out what work what does work?
12:39
SPEAKER_00: Yeah, so their core move is to build this sort of probabilistic model, right, off this otherwise really expensive operation of testing and uh and then do all the searching inside of this model.
12:52
SPEAKER_00: So the probabilistic model is a piece of software, it's also called a Gaussian process.
12:57
SPEAKER_00: And what it is is that at every um measured point, so every time you measure a point, you can add it to that, uh to that model.
13:06
SPEAKER_00: But for every unmeasured point, which you don't have an answer to, whether this when I say measured point, I mean a particular recipe, right, of these additives.
13:14
SPEAKER_00: Uh for every unmeasured point, it gives you a prediction of whether or not it's gonna work and a degree of uncertainty of how much it knows.
13:23
SPEAKER_00: And you can think about it this way the closer you are to the measured points, the more, the lesser the uncertainty and the easier the prediction it is to make, right?
13:31
SPEAKER_00: So you can be a little bit more confident about the points that are near the ones that you have measurements for.
13:36
SPEAKER_00: But at the beginning, you have no measurements.
13:38
SPEAKER_00: So all the points are equally available, which means every single design choice is possible, and everything is equally uncertain.
13:45
SPEAKER_00: So this is a big blank space you start with.
13:48
SPEAKER_00: And there are metrics, there are measures to get things going, but oftentimes there's expert knowledge, there's some um uh data that they may already have to kind of populate that initial set of dots, so to speak, in that space.
14:00
SPEAKER_00: So then once you have that, then the question is how do you get to the next step, right?
14:04
SPEAKER_00: How do you get to the next point?
14:06
SPEAKER_00: Because every time your algorithm proposes a new recipe, you have to go about testing it.
14:12
SPEAKER_00: So it better be proposing really good recipes to be tested.
14:15
SPEAKER_01: Right?
14:15
SPEAKER_01: Like you don't want it to be just a guess.
14:17
SPEAKER_01: You want it to be like an educated guess.
14:19
SPEAKER_00: Yes.
14:19
SPEAKER_00: So now let me ask you this question.
14:21
SPEAKER_00: Have you played the game Guess Who?
14:23
SPEAKER_01: Yeah, obviously.
14:24
SPEAKER_00: Yeah.
14:24
SPEAKER_00: And for our listeners who don't know that game, it's a it's a kid game where you have a bunch of pictures uh on one side where you you have this board where you can sort of pop up different characters.
14:35
SPEAKER_00: There's a grid of characters, and the characters are male, female, some have beards, some of them don't, some of them have blonde hair, brunette hair, some of them have glasses, some of them don't.
14:43
SPEAKER_00: And the idea is the other person asks questions, kind of like 20 questions, asks questions um to eliminate possibilities.
14:51
SPEAKER_00: And the smart questions are the ones that eliminate the most number of cases.
14:55
SPEAKER_00: And the, you know, you know what I mean?
14:56
SPEAKER_00: Like so a first question might be in that game might be, is your person a boy or a girl?
15:01
SPEAKER_00: Right?
15:01
SPEAKER_01: Actually, it's a yes or no.
15:02
SPEAKER_01: It's a yes or no answer.
15:04
SPEAKER_01: So you have to ask, is your person a boy?
15:05
SPEAKER_00: Oh, right, right.
15:06
SPEAKER_00: It's a yes or no game.
15:07
SPEAKER_00: That's right.
15:07
SPEAKER_00: So is your person a boy?
15:08
SPEAKER_00: And the answer to that is informative because uh potentially half the people on the grid are boys and half are girls, so you just right away eliminate half.
15:15
SPEAKER_00: But then it gets increasingly tricky because then you have to ask questions, you know, and and you can't ask very specific questions.
15:21
SPEAKER_00: You can't be like, Does your person have blue eyes?
15:24
SPEAKER_00: Well, the problem with that is if there's a lot of people left on the on the uh to be guessed, then that might just eliminate one person.
15:31
SPEAKER_00: Because the non-blue eyes includes green, brown, gray, black, all the other colors, right?
15:37
SPEAKER_01: So you have to be it's like it's smart about which questions you're asking.
15:41
SPEAKER_01: Yes.
15:42
SPEAKER_01: And the same kind of goes here when they're trying to figure out okay, what's the ratio of glucose, sucrose, L arginine, HCL, et cetera, that goes into the product that we're trying to create that will stabilize the mRNA vaccine.
15:56
SPEAKER_00: Yes.
15:56
SPEAKER_00: So the model needs to be asked the question not where is the best recipe or the best point in that space, but where is the most informative point to measure next?
16:05
SPEAKER_00: Right.
16:06
SPEAKER_00: And there's a there's a something called the acquisition function that kind of captures that aspect.
16:10
SPEAKER_00: So you want to get its predicted value, but you also want to get predicted uncertainty and you want to maximize uh getting the best, most informative answers back when you ask your questions.
16:21
SPEAKER_00: So I think that's kind of the key part here, um, is being able to kind of explore the space in a way that gets at the most informative answers.
16:30
SPEAKER_01: So even though you don't have a ton of data by using this Bayesian probabilistic whatever, you're able to ask the most appropriate questions to then feed the right things into the pipeline to test.
16:45
SPEAKER_00: Yes.
16:45
SPEAKER_00: And and they did exactly.
16:47
SPEAKER_00: And they did uh and so this is there's a mathematical formalism around Bayesian optimization that captures what I just described, kind of sort of in in general terms, right?
16:54
SPEAKER_00: But there's a formal mathematical process to go about getting at this.
16:58
SPEAKER_00: Um, and it's it's so it's under the broad subfield of multi-objective optimization, um, where you're trying to optimize many different parameters at the same time.
17:08
SPEAKER_00: Uh, and in this case, you're trying to optimize for the most information that comes back from the model.
17:13
SPEAKER_00: Um, but there's another problem, which is that let's say the model proposes one point, says, after all of this, I think this is the next experiment you guys should be doing.
17:22
SPEAKER_00: This is the recipe that you guys should be trying.
17:24
SPEAKER_00: Well, each experiment takes weeks again.
17:27
SPEAKER_00: Right.
17:27
SPEAKER_00: And a lab doesn't need to run them one at a time.
17:29
SPEAKER_00: They can run 50 plates at a time, right?
17:31
SPEAKER_00: In parallel.
17:32
SPEAKER_01: Right.
17:32
SPEAKER_01: So give me the top 50.
17:34
SPEAKER_00: So you want to propose the you want to propose 50 at once.
17:37
SPEAKER_00: Now you could take the top 50, right?
17:40
SPEAKER_00: But here's the problem with just taking the top 50 is that it'll always cluster around one promising idea.
17:45
SPEAKER_01: Oh, yeah.
17:46
SPEAKER_01: Yep.
17:46
SPEAKER_01: You want to make sure they're different in that.
17:48
SPEAKER_00: You want to make sure they're diverse enough, right?
17:50
SPEAKER_00: And so there's another technique called diversity guided selection, which they kind of carve out the space uh and deliberately spread the batch a little bit.
17:58
SPEAKER_00: Uh, and so you might not get the ideal, most optimal solutions, but you kind of cover the entire space a little bit better.
18:05
SPEAKER_00: Um, so that's another piece there.
18:07
SPEAKER_00: So then you get back 50 samples that are going to give you information, whether they fail or succeed, right?
18:13
SPEAKER_00: And that's more important than anything else.
18:15
SPEAKER_00: Um, and to find some of these candidates, they can also run some of these things on the model itself thousands of times so that they can get simulated environments, simulated evaluations to come at these uh final candidates.
18:27
SPEAKER_00: Uh, but only those final 50 or whatever gets to actually be turned into a real experiment.
18:33
SPEAKER_00: Um, in general, this falls into the family of AI algorithms called active learning.
18:38
SPEAKER_00: Um, this is the kind of type of things where the AI model, um, and I'm using AI again very broadly, it's not an LLM, it's a machine learning, statistical learning model.
18:47
SPEAKER_00: Um, but then it gets to choose its own training data in a sense, because it's deciding what to do what what next experiment to perform, right?
18:54
SPEAKER_00: Right.
18:55
SPEAKER_00: And that's going to come back as data for itself.
18:56
SPEAKER_00: So it's kind of select, it's kind of choosing the create the the best training data that's going to give it the most information to get to the best model eventually.
19:06
SPEAKER_00: Um, and and so I think that that's kind of the key here in this whole system, uh, the the whole approach here.
19:13
SPEAKER_00: And I think that it's a powerful approach.
19:14
SPEAKER_00: It's been used in other other settings too.
19:17
SPEAKER_00: Um, but like you mentioned at the beginning, it's got that generalizability aspect to it because really the recipe of four or five things could be anything and the active ingredient could be anything, and it's about, you know, it's about manufacturing and doing the right experiments.
19:31
SPEAKER_01: Right, right.
19:32
SPEAKER_01: And what's really cool is okay, they took this from theory into practice, right?
19:36
SPEAKER_01: What they did was they ended up ultimately stabilizing mRNA vaccines.
19:42
SPEAKER_01: That's incredible from this research in ways that they hadn't been able to before.
19:46
SPEAKER_00: Yeah.
19:47
SPEAKER_01: And they were able to prove that they could store it not at two degrees Celsius, not at negative 50 degrees Celsius.
19:54
SPEAKER_01: They were able to store it at 37 degrees Celsius, your body's temperature, for more than two months, and it still worked.
20:01
SPEAKER_00: Yeah.
20:02
SPEAKER_01: And you know how they told they could tell that it still worked?
20:04
SPEAKER_01: Is they gave it to rodents, they gave it to primates, and they said, Hey, are we gonna get an immune response here?
20:13
SPEAKER_01: And they were able to get an antigen-specific immune response, not inferior to what you would get with like a normal vaccine delivery.
20:22
SPEAKER_01: And this included, you know, the regular type of vaccine, but also like we talked about at the beginning, the like special patch, the one without the needles.
20:32
SPEAKER_01: And so I see this as like a really cool study because it has so many practical applications for vaccine distribution throughout the world.
20:43
SPEAKER_00: Yeah.
20:43
SPEAKER_00: Yeah, yeah, exactly.
20:44
SPEAKER_00: And this kind of squarely falls into the world of automated experimentation, right?
20:49
SPEAKER_00: So it's across science, I think, in general, this kind of approach would be powerful.
20:53
SPEAKER_01: Yeah.
20:54
SPEAKER_01: And it's like such a smart use case for AI because it's like, hey, there's a very clear problem.
20:59
SPEAKER_01: We've chosen this specific tool to help solve the problem in this context.
21:04
SPEAKER_01: And it's not an LLM, right?
21:05
SPEAKER_01: We chose a problem that was specific for this.
21:08
SPEAKER_01: We designed our experiment around it, and then we tested if it worked as intended.
21:14
SPEAKER_01: And the answer was yes.
21:15
SPEAKER_01: And now I'm curious, I'm excited to see what the next steps will be in terms of hey, can we ship vaccines to more locations?
21:23
SPEAKER_01: How much does this save us on transportation and refrigeration and ultra cooling costs?
21:29
SPEAKER_01: And also, how many more people can get this vaccine now that we have a needle less option potentially?
21:35
SPEAKER_01: And also, we may not need, you know, trained personnel to give the vaccines.
21:40
SPEAKER_00: Yes, exactly.
21:41
SPEAKER_01: Awesome.
21:42
SPEAKER_01: Well, thank you for joining us today on Code and Cure.
21:45
SPEAKER_00: Thank you.