519. Jacqueline De Lora, PhD | SURFACtoBioTech, The Max Planck Spin-Off Turning Droplet Microfluidics Into a Real-World Biotech Platform

Email: contact@bolog.io

In this episode, host Bidemi Ologunde speaks with Jacqueline De Lora, PhD, the CEO/CTO and cofounder of SURFACtoBioTech, a Max Planck spin-off developing surfactant and droplet-based technologies for more efficient, sustainable and data-rich biotechnology experiments. How can microscopic droplets function as individual test tubes? What separates useful AI-enabled laboratory automation from hype? And what does it take to transform frontier research into technology that scientists can reliably adopt? Jacqueline discusses her transition from biomedical researcher to founder, the often-overlooked role of surfactants, and the scientific and commercial challenges of building practical tools for real laboratory workflows.

Transcript

SPEAKER_01 0:15

Thanks for joining me once again on another episode of the Bid Picture Podcast. I have a special guest from over there in Germany. Over to you.

SPEAKER_03 0:28

Hi, nice, nice to uh discuss with you today, Bid. My name is Jacqueline Delora. I'm the CEO and CTO at Surfacto Biotech, a deep tech startup that is headquartered in Hamburg, Germany, that also interfaces with biotechnology. So yeah, excited to discuss with you today.

SPEAKER_01 0:50

Nice, nice. So when you first came up on my email, I was so interested in speaking with you and learning from you. I was just describing to you before we started recording how I like to learn from people, especially people that are not in my industry, not in my field. At some points in my life, I have a basic understanding of biotechnology. I know people who studied biotechnology in undergrad and master's and PhD and so on. And it's just interesting to know that now with the way technology is evolving, I want to know how you are using technology in your field. So here we are.

SPEAKER_03 1:33

Perfect. Yeah, I think biotechnology in my field, um, I'm actually really looking at it from an integration of different disciplines. So this is uh something really special, I think, about my training and my background. So I'm originally from Albuquerque, New Mexico, in the southwest of the US. Um, and I've been on a big adventure. So I live in Hamburg now, so it's a very far away to stay on the other side. Um but what was really unique about my training and my education is an emphasis on interdisciplinarity. And so from very early on, I was always interested in studying at the intersection of disciplines. I have a BS in biology, a BA in chemistry. I then went on to um study PhD my PhD in biomedical sciences with a focus in bioengineering. Um, and then my postdoctoral studies sort of launched me into a completely different field of synthetic biology. And all of this really uh culminated in the foundation for the spinoff of the company when I met my co-founder. And right now we are actually working in a way to integrate all of these sort of like disparate disciplines in a way that is positioned to enable the future of biology. And what when what I mean um when I say the future of biology is really generative biology. Um, and so this is really mirroring the current state in AI as well. We have this idea of generative AI, and it's already being applied to biological models. Um, but in the field in biotechnology and especially in deep technology, where we are building our innovations, it's really towards the interface and towards developing biology and sort of like the next round of evolution in you, if you will.

SPEAKER_01 3:52

Wow, wow. I like how you describe it that way because from my most basic knowledge of biology, if I still remember some 30-something years ago, is evolution happens iteratively. So mouse or rats, I guess they learned how to climb on trees, and then they realized if they jump from one branch to another, they can get more fruit and nuts, and then they evolved wings, and then they became birds. I'm describing this from a very rudimentary level, so please forgive me. And birds eventually found out okay, if they evolved their wings, then they can fly farther and higher, and then eagles became a thing, and then eagles evolved, and then they became something else, and then fish evolved, and then they became sharks, and so on and so on. My listeners who have much more knowledge are already type composing their emails. Forgive me again, forgive me, I don't know anything at all.

SPEAKER_03 4:53

So no, I think you're totally right though. You're tapping into Darwinian evolution exactly. Yeah, and in history, um, in the history of nature, if you will, Darwinian evolution takes a lot of time, actually. You know, we're talking uh about million eons really of spans of time in order for these sort of like features or fitness um phenotypes to evolve in different species over a specific amount of time. And now what we're able to do in the lab with biotechnology is accelerate those timelines and actually on a certain level take the idea of evolution into the hands, into the test tube, and actually start to design biology in a way that um it can be useful in therapeutic interventions, for example.

SPEAKER_01 5:55

So thinking of therapeutic interventions, um, what does surfacto biotech do? What are you trying to build and why is it important for folks like me who don't know anything about biotech?

SPEAKER_03 6:09

Yeah, uh maybe I step one moment back towards when you use the word iterative or iteration. Um in pharma and in biotech, the way workflows are operating in these spaces, um, it's called DBTL. And this stands for design, build, test, learn. And this is a cycle that people in the lab can use in order to pose a hypothesis or a question, um, design everything that you need around that question, and usually nowadays that is actually happening in silico or using computers, using generative AI. You can actually, for example, predict the three-dimensional structure of a protein. And drug designers, medicinal chemists, if you will, they are directly responsible for designing how drugs can interact with proteins. And when those types of lock and key mechanisms are well designed and well tested, it leads to the unlocking of therapeutic compounds, classes of therapeutic compounds, even, that can uh lead to the solving of global health issues that range from things like HPV all the way to cancer, all the way to different types of um global health epidemics. And so it's really already valuable that we have these AI predictive models in the design layer of this DBTL cycle. But where my company is stepping in is in the next two layers, in the build and the test. This is the literal implementation in the physical world. So AI can design today, it's something like 200 million proteins have been, uh protein 3D structures have been designed and predicted using models on the scale of a program that was brought out by DeepMind, um, which is from Google, uh, called Alpha Fold. But the problem is that those over 200 million proteins, they're just a prediction, they're just an estimation. They haven't actually been experimentally validated. And with the current technology, we're already really far. So don't get me wrong. We already have automation, we already have robotics, we have high-content screening centers that aim to be able to understand better how protein drug interactions are working directly. But what we recognize with uh surfactobiotech is that it has already reached its limit. And there needs to be a disruptive innovation in the way these protein compound interactions are actually working in the lab. We actually need the experimental validation in order to really understand these things better so that we can achieve the learned layer of that cycle and go back into the next round of design knowing more and knowing more from the physical reality, not just from the virtual space. And so we do that build test layer by literally disrupting the test tube. So we have this really cool technology that is called droplet-based microfluidics. So I know that's a really deep, long term. Um, but it's something that I have spent the last 20 years obsessing over and studying, and I'm right now with my co-founders working to bring this technology into this space so that it could really make a positive impact on that build test layer in DBTL. And essentially what we're able to do is change the value of the biological sample that gets processed from the container that is the test tube. And so you can sort of um we can make some general assumptions. We can say that one test tube usually has a couple milliliters of biological sample. And depending on the question or the hypothesis that you're trying to address, that biological sample can be very expensive to produce in the first place. And then we can just generalize it and say that every test tube can only give you one data point, right? And so that establishes the economic value of running a lab test in a test tube. And what my technology is able to do with this droplet-based microfluidics is to take a very small volume of biological sample where the basis of the sample is a water-based environment. All biology happens in water. It's a really cool feature of nature, and we can manipulate this feature of nature by using tiny engineered channels that are about the size of a hair in a chip. And this chip architecture, we build it in a way that there is a T junction, and in the central channel of this chip, we can pump in the water phase that has the biological reaction. And then in that T junction, in the cross-cutting channels, we can pump in an oil phase. And because oil and water do not mix, when those, yeah, when those two phases interface within these microchannels, they actually form droplets based off of differential pressure, and you can also add some mechanical features on top. And so we're able to produce large populations of tiny droplets. They can even encapsulate one human cell, maybe even multiple human cells. Um, and this enables us to make these compartmentalized water environments in an oil solvent. And now with only 10 microliters, a very tiny volume of the fraction that you normally need in a test tube, we can produce hundreds of thousands of these droplets. Each and every one of those droplets is the unique source of a data point. And now the value of my test tube, I collect those droplets directly in the test tube, actually. And so at the end, the value of my test tube, um, sort of an economies of scale is applied to this test tube. And I'm actually able to invert the cost to data point ratio because with so much less sample, I'm able to harness so much more information. And so this literally breaks the physical scaling limitations that we have from just deciding to use the test tube as the container in the first place. So our droplets are analogous to a test tube, just drastically smaller, orders of magnitude smaller each, but running in massive parallel experimentation platform. And so this unlocks the build test layer of the design build test learn that enables us to literally speed up testing, accelerate testing, and to run campaigns that would normally take weeks to months worth of work, now down in days instead.

SPEAKER_01 14:39

So the build test layer can be applied to pretty much any biological sample. Because at the very fundamental level, all biology is based on water.

SPEAKER_03 14:53

Yeah, exactly.

SPEAKER_01 14:54

Like blood, um, plasma.

SPEAKER_03 14:59

Even tissue-based uh samples, uh, it's possible to process them there. That's actually what my PhD thesis was focused on. So, from the like I said, from the very beginning, I've been passionate about droplet-based technologies. And in my PhD, I actually developed methodologies to encapsulate multiple cancer cells into droplets and use that encapsulation to form essentially small models of tumors in a test tube so that we could, first of all, better understand the microenvironment that the cells are living in. So we could ask questions about oxygen, pH, very just fundamental chemical microenvironment types of um characterizations. And then we could extend that to how do those cells respond to chemotherapeutic interventions on an oxygen gradient, for example. And this starts to lead towards understanding where chemotherapeutic resistance comes from, because typically in a growing tumor mass, you have gradients of oxygen, you have gradients of pH, and you also have gradients of waste products from metabolism like lactic acid. And that changes the phenotypic expression of those cells because they're responding to their environment very much in a Darwinian evolution type of gene expression change. Um and you can use droplet technology to increase the resolution on these types of correlations and measurements to just better understand the biological system from the beginning. Um, maybe just one thing that I really need to mention as well, where my company name is coming from, it's super nerdy. So we're called Surfacto Biotech. And in this uh production of these water droplets in oil, if you really did it like this, um, and you you made your droplets in your microfluidic channels without a very special ingredient, what would actually happen is the droplets bump into each other and they merge. They do a phenomenon called coalescence. And that causes a bulk phase separation. And so you would end up with the water phase and the oil phase separated. Much like when you make up a salad dressing, you shake it, you use it, but maybe not all of it, and you place it back into your cabinet or into the refrigerator, you come back the next day and it's separated. You have to shake it again. So the way to fix that is using a cool molecule that is called a surfactant. Surfactants, yeah, surfactants. Surfactobiotech. Now you get it. Surfactants for biotech. This is really where we're coming from. Um, and and it's coming from my co-founder. His name is Martin. He has a PhD in chemistry from the University of Heidelberg. I met him when I very first moved to Germany as a postdoc in synthetic biology in the Max Planck Institute for Medical Research. Uh, we can get into that if you have questions, but essentially, he is a mastermind at building surfactants, at the chemical synthesis of surfactants. And essentially, what these surfactants do, they are a molecule that has two phenomena of nature built into one. Part of the molecule is hydrophilic, meaning it loves orientating towards water, and the other part of the molecule is hydrophobic, meaning it hates water. It hates water, and therefore it loves to orientate towards oil. And so you can imagine when we create this forced interface within the microfluidic channels where we have these water bubbles surrounded by oil, well, the surfactants come in and they self-assemble at the interface. They form a molecular layer between the water compartment and the surrounding oil phase, and they essentially form a shell that makes it so that when the droplets bump into each other, they can't coalesce. They're stabilized against each other. And this is exactly the feature that unlocks our ability to use these droplet micro compartments or micro test tubes over time in the lab to uh ask questions about biology.

SPEAKER_01 19:43

Wow, wow, this is so cool. And like you alluded to, I have so many questions, and you mentioned when you moved to Germany. Okay, let's rewind all the way back to New Mexico.

SPEAKER_02 19:56

Sure.

SPEAKER_01 19:57

In New Mexico, and me personally, I've been to all 50, I've traveled to all 50 states in the US. It's just a personal goal I've had. Thank you. Personal goal I had when actually Colorado, Arizona, and Florida. Those are my top three.

SPEAKER_02 20:13

Okay, okay.

SPEAKER_01 20:15

So I made it a point of duty to visit all 50 states, and I drove through Albuquerque and Santa Fe. And I was going north to south, so Albuquerque, then Santa Fe, and then I ended up in El Paso, Texas, you know. So but New Mexico and green chilies. Don't even get me started. That's an another episode entirely. Anyway, we can do it.

SPEAKER_03 20:41

We can do it. I would be happy to talk with you for an hour about green chili. Oh, yeah. It's it's a passion as well.

SPEAKER_01 20:48

Yeah, it's it's like a I don't like it. It's a state pepper at this point. They have it on their license plates. Every car's license plate has green chili plates.

SPEAKER_03 20:57

The state question question is actually red or green, uh, because you can have red and green chili. And uh the answer to this question as a local is Christmas, because you can order both red and green on your uh huevos rancheros, your burrito, whatever. And I can really suggest it. Anyone who passes through New Mexico, the answer to the question is Christmas, and yeah.

SPEAKER_01 21:26

That means you you really you really know what you're talking about when you say like, wait, what? Red, green, okay, pig green? No, this is it for you.

SPEAKER_03 21:33

You don't have to choose. You can have both. It's amazing.

unknown 21:38

Yeah.

SPEAKER_01 21:39

So so you grew up in New Mexico and then you moved into biomedical science and microfluidics and Max Planck research. What early experience made science feel like where you're most comfortable? Was it in high school? Was it in college?

SPEAKER_03 22:01

It was actually way earlier than high school. Um, one of my most, I mean, there were a couple formative um experiences for me. Very early on, um, I was sort of lucky to be exposed to a pipeline that was really focused on getting girls into STEM fields, into math, into science, interested in these types of disciplines. Um, and so I I was encouraged from a very early age to design science fair projects. You know, I don't know if you remember, you have these cart cardboard sort of like three, yeah, three surface uh, you know, you have to print out your your slide design and paste it onto this board. And uh I mean, I was doing that very exactly. Yes. Thank you for the display. Exactly like this. Yeah, posters, you know, you learn you learn early how to explain your technology or whatever to uh to an audience. And I was doing that. And really like fifth grade is where it was at for me, actually. So in fifth grade, I posed the question what is the difference between music and noise? Um, and I was just, you know, really it actually coming from from my mom and my dad, sort of like a combination of their personal interests. My mom was really into ballet, my dad was really into rock music. And, you know, we always had music playing in the house, either um for a fun dance party situation or for classical music while we're studying kind of thing. Um, and my dad had this really cool advanced speaker system where um there was sort of like there was a feedback output, and I could see what this what the sound waves were actually looking like from the music that was playing. And I started to notice that it's different from classical music to jazz to um the Beatles to whatever was playing, right? And so I there was like a fundamental question that was always really interesting to me. And in fifth grade, I posed that as my science fair question. Um and I that actually created a thread throughout my entire career. And now I literally have patents based off of the use of sound to manipulate fluids, to create water droplets, um, to create controllable water droplets. So there's sort of like this yeah, overarching theme that started from a very early age. Um, and the other really cool thing that I did that early um in fifth grade, we had an opportunity in my class to join something called Mars Project. It was run by um the national labs that exists in New Mexico. There are two there. And um, the idea was to essentially build a colony of pods that were connected that would simulate the ability for humans to live on Mars. And I was the project lead in fifth grade for this, and I got to go to a really cool telescope out in the middle of nowhere in New Mexico, and I got to look into the universe, and you know, it's just experiences like that, being able to ask my own questions, being able to have an exposure to some of the highest technology that existed at that time, it really was formative for me. And and so, although it took me a while to come back around to science by the time I was in university, um, it really just made sense for me to continue like this.

SPEAKER_01 25:53

Nice, nice, wow. Uh for my own background, as more of so my both my parents are academics, they're both professors, and it was kind of like an automatic path for me and my siblings to end up in academia. And growing up, I've always been interested in just technology and how electronic devices work and the video game over there, what makes it possible for me to take the controller? And I know there is a connection, the controller and the cable to the console to the TV, and then there's this big giant feedback loop. And I'll just take a screwdriver and you know, curiosity. What is making this video cassette recorder? Now I'm dating myself, but anyway, what's making this video cassette recorder play these cassettes with the magnetic tape? And sometimes the thing looks like it's dirty in code, and I'm just like, what makes it dirty? The dust getting there, and then they teach us how to use um wipes, a cloth, silky cloth, and some disinfectants, and you can't clean it too much, or you damage the clayhead and all this cool stuff. And that's kind of you know, growing up, I was like, okay, well, I have to study electrical engineering so that I can satisfy this curiosity, and so on, and so on. And along the line, I realized, okay, well, it's one thing to learn in school, but ultimately you need to learn about the world you live in because I'm a fundamentally curious person. Have you ever noticed how driving on the road you notice some parts of the road have metallic lampposts, and some parts of the road have wooden lampposts, yeah, same thing for electric poles and telephone poles, and I'm thinking, why? The same stretch of road, metallic, metallic, wooden, wooden, and then you come across concrete lampposts, and I'm just like, why is that? So those are the kind of things I'm okay. Well, I can find out by talking to people, engineers. This was way before Google became a thing, and now I can like satisfy my curiosity. So now I have a seven-year-old boy with it's he's kind of like that as well. And I just tell him, Okay, well, some questions I can answer, some questions remind me when we get home, we're gonna go check on Google. And now he's asking me questions like, Okay, this bird has this type of wing, this other bird has this other type of wing. And he's like, Why? These are the kind of questions I would also ask. Shouldn't all birds have the same types of wings? Apparently, not because even birds of the same size have different types of wings for different reasons, yeah. Evolution, right?

SPEAKER_03 28:48

Yeah, exactly. We're all a function of our environment. Yep, yep, yeah, wow.

SPEAKER_01 28:54

Speaking of environments, um, your PhD research was on 3D cell cultures for analyzing tumor microenvironments. See, I did my homework.

SPEAKER_03 29:09

Well done.

SPEAKER_01 29:10

What made thank you. What made you decide to look into the microenvironment that a tumor grows in? I know you touched on that earlier, but just to you know expand a little bit on that. Um, what what was that thing that made you curious about microenvironments for tumors?

SPEAKER_03 29:28

Yeah, I mean uh it's uh a little bit of um a personal motivation, first and foremost. So uh cancer has touched the um my my family essentially in many different ways. Um, from a very early age. I lost my grandfather to a brain tumor. Um and then fast forward into later in life, and my my one of my very best friends had um a super devastating breast cancer diagnosis. Um my husband had skin cancer. Um there's just you know, the cancer, it's it is really one of the most devastating types of diseases, no matter the level of cancer that you get. Um it's uh it is a global health issue that I have seen directly firsthand the implications and what it brings to the family dynamic, to the healthcare system, just the entire stress that the um development and evolution of this multifaceted disease um brings brings to humanity, brings to like life science, essentially. Um and so in my PhD studies, it was really natural to just, I think first of all, I started from a question more towards what is life? This was like one of those early questions that I was also asking myself, like, yeah, why are the birds all different from each other? But also how are they alive in the first place? Like that that was I was going, you know, really philosophical foundational questions from a very early time. And by the time I was in my PhD studies and I had the opportunity to go into courses that really focused on cancer biology, I really started to understand from a fundamental level that everything that I had learned in my foundational biology courses could be applied to human health and disease, and especially in cancer, and especially when you start to try to simplify the purpose of a cell within its microenvironment and what happens when it becomes dysregulated or knocked out of homeostasis for whatever reason. There's this question how does how does a cell become a cancer cell in the first place? Um, and I really started to become obsessed with that question, and I wanted to know more about how the microenvironmental conditions impacted that directly. And at my university in New Mexico, there was already a lot of research going on around toxicology, around um uh sort of like metabolics, uh, how what we're eating is implement is implemented in the development of cancer, how the quality of our air is implemented in lung cancer, types, things like this. But I really wanted to zoom in to the tissue microenvironment and understand on those tiny scales what is actually happening, what is developing over time that is leading towards, first of all, a cell to become canceric in the first place. Second of all, how the microenvironment is implemented in this uh transition. And then also, what I then realized is that the tumor environment is completely heterogeneous. And so we are facing a super deep problem with respect to cancer and our the technological development that needs to happen in order to solve or cure cancer, if you will, um we're we're still working on it as um as a human race, right? And one question that that most of my my question, what of life what is life, has sort of evolved into okay, well, we have AI now. Why hasn't cancer been solved? Why haven't we been able to ask the artificial intelligence to propose a solution to this multifaceted problem that we face as humans? And what I all what I really come back to at the end of the day is we actually just don't have enough real-world validated data for the AI to make the proper predictions to assist us in having a solution to something like cancer so far. Um, and so it's really a personal mission of mine within the context of Surfacto Biotech to become an enabling layer in those DBTL cycles so that we can produce enough data to be able to answer questions like this that span from what is life from a synthetic biology point of view all the way towards the therapeutic interventions.

SPEAKER_01 34:29

Wow. Wow. Thank you so much for sharing that. And sorry to hear about all the different diagnoses that nobody wants to hear about. And I hope your husband is doing well.

SPEAKER_03 34:44

Yeah, everyone is okay. I mean, my grandfather passed away early, but that was a very long time ago. So um modern medicine is, you know, it is working. There are therapeutic interventions that can come, but they come with a cost. They come with scars, they come with um um body dysmorphia, they come with, you know, a lot of um just like psychological factors essentially. And so just for the overall human condition, it would be really amazing to just make small improvements, even if they're incremental, they s they still need to come so that we can really advance, advance forward.

unknown 35:21

Yeah.

SPEAKER_01 35:21

Right, right. So why did you decide to um basically create a startup? Because from my limited knowledge of the scientific world, most innovations begin as an academic research and then they publish a bunch of papers and they present at a bunch of fancy conferences, and you know, you know how it is.

SPEAKER_02 35:44

Yeah.

SPEAKER_01 35:45

But you decided to to go the startup route. What convinced you that this needed to be a company that would make products?

SPEAKER_03 35:58

Yeah, so I mean, it was sort of multi-sort of like two things. Um, first of all, it was a function of um the way the current academia is working um on a worldwide basis. So, you know, I I sort of did this crazy thing where I went from the US to a foreign country in my postdoctoral studies, had a very fruitful five years in my uh postdoc, and then started around the last one and a half uh years of that time to question what's my next step? What should I be preparing for next? Because the postdoc, usually, if you um stick with the sort of like classical trajectory, the postdoc should lead into a professorship. And so then I sort of like did the same thing that I did after I finished my PhD studies, and I started to look internationally to see what potential academic department or university around the world would be able to um match me, essentially, would be able to offer me a place where I can bring my super interdisciplinary background in order to start to really build the next layers and to also give back in a way that I'm training the next generation of scientists. And when I went um into this search, if you will, I actually just came up empty-handed. I was trying really hard to make puzzle pieces that were just not engineered to fit together, fit together. And there was a lot of friction. And at one point, I really just had to sort of like sit back and make make a little bit of room for reflection. And within that time for reflection, I also started to lean on the people that I had around me. And one of those people was Merton, my co-founder. Um, he was also within the same group at the Max Planck, and we were working together all of the time, late nights in the lab, overnights in the lab, really pushing forward on a really cool field in synthetic biology, what is to the idea is to build a synthetic cell. So to build a cell out of molecular assemblies such that it starts to mimic a natural cell. Because when you can reverse engineer something, you can really fundamentally understand how it's working, like you, when you're taking apart electronic devices to see what's on the inside, to know what it's made out of, to see how it's functioning. That's the same idea in this field of synthetic biology. It's the niche field in synthetic biology. Um, but, anyways, Martin and I were working around projects and problems like this, and I just expressed to him that I was really frustrated that I was not going to find my place in the world. And he had a similar feeling as well. And he also had been working on developing surfactants, and we were coming together and we were merging our expertise, and we were actually sort of discovering that us together was producing something really unique and really useful. And we we actually have an elevator story, so I can tell you, I can tell you, it's not an elevator pitch, it's an elevator story. Yes, please go ahead. And so one of these late nights, um our labs were in the basement of the Max Planck where all the physics labs are, right? It's dark, it's stable, it's a quiet environment, and so we were doing experiments there. And we were on the way up to our offices in the sixth floor, and we're riding the elevator together. And I'm just telling Martin, oh man, Martin, I have such a problem with stabilizing my droplets. I have these commercially available surfactants, and I tried them all, and none of them are working for this really unique and emergent type of droplet that I was working on. And I know that he has a really good reputation within our group as being the mastermind surfactant synthesizer, like I mentioned before. So I asked him, Hey, here's my phase orientation. Do you think you could come up with something? And he's a yes person like me. So he's like, yeah, maybe. And you know, that's the end of the elevator ride. Okay, he goes away, comes back to me a couple weeks later, and he hands me a vial of surfactant of the product, and that unlocked my field of research. That unlocked the this crazy idea that I had been working on that I had no commercially viable source. Um, and so, sort of like it's it's um an assembly of these things where we had an eye-level understanding of wanting to be able to create a space in this world to work in a way that we can actually bring our skill sets and our technology as an enabling layer for all of the other researchers, lab tests, industrial biological data production, if you will, to really enable that in the first place. And so um, yeah, Martin originally had this idea for the startup, but because of these foundational and sort of nucleation interactions that we had together when he asked me, I also had no hesitation. I also jumped into my yes mode. And ever since then, we have just been working to make it a reality. And so now I created my own position. I didn't, I didn't have to make the friction, you know, I didn't have to overcome the friction. I actually found a different way around it. And so, although it's not in the academic space, we still give back to the community. Um, we still uh donate our time, essentially, we volunteer our time towards the education of students working at the interface between scientific fields and entrepreneurship, because this is, in my opinion, really where impact can happen. Impact really needs to come from a few different places. It needs to be economic for sure. It needs to be um impact for for life, for the greater understanding of life. And when we talk about therapeutic interventions, the biggest impact that you can have is on the patient's life at the end of the day. And so we now created um essentially a mechanism in order to create a reality around that for ourselves.

SPEAKER_01 42:54

Wow. Wow, thank you again for sharing that. I like hearing founder stories, and it just goes to show that it takes a multidisciplinary approach to solve unique problems. And uh the more people I talk to, the more I start to realize this is not just within science or engineering or cybersecurity, it applies to pretty much anybody in any phase, even elementary school kids. You have a unique problem, some kid is not trying to share their toys with you. Well, if you approach it from a five-year-old's multidisciplinary approach, you can actually end up making more friends by the end of that school year. I've seen that in my own son, but I digress.

SPEAKER_03 43:46

It's totally right, it's foundational human behavior. Yeah. Yeah.

SPEAKER_01 43:52

So you you moved from Heidelberg to Hamburg. Um the way my brain works, maybe you were not expecting this question.

SPEAKER_02 44:01

Okay.

SPEAKER_01 44:01

I do my homework and I pretty much ask why for everything. So, why did you move from Heidelberg to Hamburg?

SPEAKER_03 44:12

I mean, many people have asked me this question. Um, and as a very simple answer, I think. We were essentially okay, so we had already developed a really deep network in Heidelberg in the south of Germany. Um, it's also a biotechnology hot. Spot there. There is so much amazing academic as well as entrepreneurial work going on there in the biotech space. And we had a network there to a certain extent already. And at the same time, we started to look for lab space. And lab space for a deep tech startup is an essential. It's not an option. And to find lab space like openly available existing lab infrastructure is a big challenge, especially in Germany. And so when we recognized that we we didn't have space in the Max Planck anymore and that we really needed to sort of, you know, lead the nest, spread the wings, get out, if you will. We started to look around everywhere. We did a global search for all of the sort of biotech, deep tech hot spots. And we wanted to find a place where we could, first of all, have access to laboratory infrastructure. And second of all, open up into a brand new network. We really wanted to have the opportunity to expand. And so there are, there was a lot of contenders, right? We looked everywhere. And there are a lot of amazing cities working on really cool concepts to support entrepreneurs, especially in deep tech and life science. And we came across Hamburg, and Martin actually had already a very good picture of Hamburg. I had never been there before at the time. And so we reached out to the existing startup support that Hamburg already had, and we recognized that they were working on this really cool urban development project where it's called Science City. And the idea is to create one campus that integrates startups, academia, and industry in one co-localized area to just accelerate progress. And at the time, they had been working on building the very first building that was the nucleation point for this campus. And we ended up being one of the very first, the first tenant actually in that building. We built up our lab space. We moved there in September of 2024. It was a brand new building. And so we yeah, it took around a year in order to establish our facility. We have the ability now, we have a biosecurity clearance, so we can do genetic engineering work. We also have a full-blown chemistry facility. And we essentially built our lab from two empty rooms into a state-of-the-art droplet, droplet lab. And yeah, so essentially, long story short, we needed lab space and we needed to expand the network. And Hamburg made an offering that we couldn't resist. And if I'm not sure if you've been to Hamburg before, but it's a port city. And it's really the city in Germany that opens up to the world directly from the way that it's built. And I have I just continue to fall in love with Hamburg. It is my second home. Albuquerque is always first. Um but Hamburg, Hamburg took second place, and it's uh it's a really cool international vibe. Um, where I would say the startup ecosystem, I am at the very beginning, I'm starting to see the bloom of the flower start. And it's uh it's really cool to be a very first mover and shaker in that type of ecosystem.

SPEAKER_01 48:19

Nice, nice. Well so um in your world, where do you feel AI can genuinely help like droplet microfluidics?

SPEAKER_02 48:36

Yeah, that's a good question.

SPEAKER_03 48:40

Oh, I really think that it's um it's a combinatorial effort. I think that we as um entrepreneurial scientists uh from surfacto biotech side, what we're really working on doing is just improving the quality of data that is leading towards building machine learning algorithms that at the end of the day serve the AI to have less hallucinations, to have better data input, so that what it potentially can propose as the next iteration in these DBTL cycles is actually grounded in fact, in reality, and and really pulled down out of the cloud, out of the virtual, into something that is truly tangible and truly impactful. Um, and I think that really we need to focus on generating the data that AI needs in order to actually be less artificially intelligent and more real-world intelligent. Um, I really think that, you know, it's it's sort of the cat is out of the bag already. There's really no going back from AI. It's there, everyone is using it. Um, and I am a little, I have a little bit of worry about it with respect to human creativity and human intelligence as well. And so I think it's really important that we step away from the virtual often. We put our hands onto real-world biology, real-world um mechanical engineering, electrical engineering. We don't forget that the fundamentals actually happen in reality. And we really um yeah, just improve the way that those that that the reality is translated into the virtual space.

SPEAKER_01 50:44

I read a paper, um, I want to say early this year about how AI is enhancing drug discovery. So again, the way my brain works, I see something, I become interested, I read tons about it, starting from something I find on Google Scholar, as one does, to picking up a random magazine. Maybe I come across a bookstore at the airport and I see a magazine and it talks about something I've been interested in, and then I read it right there. I'm not buying the magazine because I don't need to. I just need to read one article in one magazine at a random airport in Midwestern US. Anyway, I was reading about drug discovery and how AI is making drug discovery faster because the way pharmaceutical companies work, they have to do a bunch of testings and clinic clinical trials and this and that. But AI models, they now build AI models to make it faster that okay, well, this drug can work on this particular group of people with this particular genotype and blood type and so on. So we can skip like six different phases and then just go directly to making the drug. And I see it a lot in cybersecurity. Actually, you're trying to get someone to not hack your company and you have all this data, but then someone can be, you know, poking at your defense at your firewalls, and you wouldn't be able to detect it. But now you put an AI system in front of all that data, and the AI system can say, wait a second, this traffic from this particular IP address looks like someone is trying to hack you, they're checking out where you're the weakest link in your network is. Pay attention to this one, just block that IP address and your problem is solved. So a lot of that is you know beginning to show up. And like you described, it's important to pay attention to the physical system before we can then layer AI in any direction we want to layer AI. Maybe find out how something works within a particular cell system or a cell structure or da da da da. And then you can now have a full understanding of okay, well, if we use AI at this phase, it's more efficient than using AI at the very beginning. Because a lot of people just want to patch AI at the very beginning of their workflow and assume it's going to make everything more efficient. Not necessarily. But what do I know?

SPEAKER_03 53:20

No, I really I think that's exactly right. Um, I and it made me start to start start to think about, you know, with respect to like cybersecurity and monitoring. Well, cells actually are doing those types of functions. You know, there's always processes happening in cells. We call it cell signaling, right? There's always like constant turnover of protein interactions. And you can you can throw the wrench in the will with a drug, actually. And those drugs can actually modulate exactly how cell signaling and cell function is working. And yeah, I mean, maybe one day it's looking like we have an AI-generated model of cellular function that we've built from a bottom-up approach because over since the 1970s, essentially, we have been studying and populating databases with the information in order to understand how cells function. And when you know how a cell is functioning in normal conditions, then you can also start to ask questions about why it's functioning the way it is in abnormal conditions. And perhaps AI surveillance becomes sort of like a biointerface, if you will. And it we could maybe eventually use it, like you're saying, not at the beginning of these ZBTL cycles, but really harness it at the learn level so that we can, yeah, design better therapeutics, even on a deeper level, just reach the understanding of why, why abnormalities, why diseases are emerging in the first place. And I mean, who knows? I think that when the human condition is improved, that can only extrapolate to the global condition as well. And so it's it's really motivating to work on human health and disease. And let's see how AI gets integrated. I can only main maintain optimism for it.

SPEAKER_01 55:16

Right, right. Kind of like how AI shouldn't replace the experiment, but it should help the experiment correct itself. So while preparing for this um conversation, I was reading how AI-assisted digital microfluidics research is showing that AI is now being used for droplet recognition, um, feedback control, automated manipulation.

SPEAKER_02 55:42

Yeah. We're building this in our lab. Yeah, yeah.

SPEAKER_01 55:47

Wow. So it's pretty much here already, the whole AI revolution helping everybody.

SPEAKER_03 55:54

Yeah, we're at we're at the foundational layers for sure, but this is exactly uh something that I'm working in the lab with my RD team uh constantly, is how can we build um machine vision feedback control in our hardware devices? Um, really, it's towards improving the life of the scientists and the technicians standing at the bench so that they are freed up to really scrutinize the data that is coming out of the experiments. That's that's really the most important part. That's where human intelligence is needed the most, is in the analytic layer, right? We don't really want to outsource that to the AI because, yeah, the code needs to be checked essentially. And so it's actually, it's maybe better if AI doesn't generate the first draft of the code, but rather the first draft of the understanding is coming from the human layer, first and foremost, and uh to implement the AI in a way that it makes our lives easier to get to that layer in the first place. That's uh sort of how we're thinking about it. Um, but yeah, droplet recognition, droplet feedback control, all of these things are are becoming a reality. And it's it's super cool to be working on the frontier of that. Yeah.

SPEAKER_01 57:10

Nice. And I saw something on your website that says you're looking for shadow testers.

SPEAKER_03 57:18

Yeah.

SPEAKER_01 57:19

What's what's that all about?

SPEAKER_03 57:22

Yeah, I mean, it's maybe the a better terminology is beta testers.

SPEAKER_01 57:26

So we are fancy word.

SPEAKER_03 57:29

Yeah, yeah. We already have products that we have produced in our lab. We've already sold surfactants um essentially to academic researchers. These are the guys who already are trained in producing droplets, and they actually, much like me a couple years ago, just need a better surfactant. Um, and so we're able to, at the beginning of this year, we were able to produce our first batch of uh our surfactant products, and those are in the hands of some of our customers already. And in addition, we are also designing best in-class droplet production devices as well. And so these are sort of like consumable aspects that we are already trying to get into the market so that we can get early customer feedback so that our own DB tail cycles for product production can make sure that we are designing the products in a way that our end users find them useful, actually. And so we're really working on product market fit from the very beginning. And the shadow testing programs or the beta testers is how we're doing this. So you get a nice discount, and yeah, anybody can go to the website and reach out to us on the forum. And yeah, we're we're really happy to um yeah, get our products into the hands of our users and also work on code developments and pilot projects as well. All of these things are um within the docket of how we're working at Surfacto.

SPEAKER_01 58:55

Nice, nice. So um when a Bell Tech team looks at a new Microfluidics platform, what should they be most worried about? Is it reproducibility, integration, regulatory uncertainty, cost training, or something else?

SPEAKER_03 59:15

I think that it's mostly about adoption into existing workflows because you really don't want to be disruptive, especially with biology. A lot of the times it takes years in order to establish your working protocol. And so when you have your working protocol, you're very resistant towards introducing any new technology into it. Um, and so the adoption layer, I think, is the biggest hurdle that we face with droplet technologies. And it's because the technology in the past sort of developed um a tricky reputation, if you will. It's not exactly easy with the current status quo of the technology to implement it into existing workflows. Um, and so this is something that we are exactly working to solve with our with our solutions at Surfacto. We're trying to build in a way that the adoption barrier is drastically reduced so that it's very simple, automated, and easy to implement it into your workflow. And actually, the the platform sits at the beginning of the workflow, right? Because, like normally you have your pipette as a biologist and you're transferring fluids around in a well plate or a test tube or a few test tubes, and you do this manual work. That manual work has also been automated with robotics, of course. Um, but at the end of the day, it's really about fluid handling and liquid transfer. Um, and so, like I said, at the very beginning, we're disrupting the test tube. And so, in order to disrupt the test tube, it means that you need to plug in one extra step in your workflow at the very beginning, so that you can create the test tube environment in the droplet sense now. And that unlocks an inversion of the cost to data point ratio, first of all. So now you can get a lot more data for a lot cheaper, and that overall just leads to the acceleration of your time to result. And that can be applied definitely in research use only cases. This is where we're working first because the regulatory hurdles are lower, but we have big aspirations to go into regulated markets and spaces because we see that that is where that impact layer for human health and for patient for patient impact comes.

SPEAKER_01 1:01:41

Nice, nice. And I'm keeping one eye on the clock. I know it's late for you.

SPEAKER_03 1:01:47

No problem.

SPEAKER_01 1:01:49

Final question before we begin to wrap up. 10 years from now, in the year 2036, when you look back at all the cool stuff you're working on now, what would make you say surfacto biotech didn't just make a big impact in the industry, but you changed the way biotech experiments are done.

SPEAKER_03 1:02:17

I think for me, um the easiest way to sort of like measure that level of progress progress would really be to take it to the clinic and to say that um, especially with respect to breast cancer, this this is one of those cancers that uh really makes me tick. Um to say that the cues in hospital diagnostic centers are low to even non existent, because the droplet-based technology enables diagnostics on a level in scale and accelerated pace such that patients don't have to wait for the answer to diagnostic test results before their therapeutic interventions can be designed by a doctor. Um, and so I'm really talking about, you know, genetic test results, for example, coming from a blood sample. I don't understand why it takes weeks sometimes for these test results to come back. Those weeks uh are full of existential dread for the patients. They don't know what their what their the therapeutic outcome looks like, they don't know how long they're gonna survive. They're waiting to have the full picture of their diagnostic in order to understand what is coming for them. And I think that that level of stress is is counterproductive to the overall um sort of mission of trying to help these people in the first place. And so when when I could say that droplet-based technologies could deliver a diagnostic result in the very same doctor's appointment where biological biological sample is collected in the first place, that would be a milestone where we made an impact, not only economically speaking, but impact speaking.

SPEAKER_01 1:04:17

Wow, wow, that's quite powerful. Thank you. And one final, final question.

SPEAKER_03 1:04:24

Okay.

SPEAKER_01 1:04:26

For young scientists who are listening, um, think about yourself in let's say middle school. What should they learn now if they want to build something cool and impactful that is multidisciplinary and it's one leg in science, one leg in AI, and something that can be adopted in the real world with measurable impact. What should those middle schoolers start paying attention to now in middle school?

SPEAKER_03 1:04:59

Math. Math class is where it's at, statistics, really get gaining an early understanding um towards these fundamental descriptions of the world that are based in numbers, first and foremost, um and moving that towards calculus as well. Um it's it's really I think that this is also deeply related to AI. AI can solve equations, AI can formulate equations. Um and so, in order to understand how AI is working in the first place, how code is working in the first place, the fundamental goes all the way back to mathematics. And so for the for the future generation, not only is about being able to formulate your own thoughts and communicate and write your own thoughts from a place where AI didn't impact it in the future, in the in the first place, but also to understand the mathematical implementation. Um, because it's it's almost one of the most beautiful ways, other than words, to describe the physical reality. Mathematics was really like the very first version of AI. And so I think uh math is where it's at. And you know, it's uh it's a hard subject, it can really challenge you, but when you gain the mastery of it, the world unlocks itself. And so, yeah, pay attention to your mathematics professors.

SPEAKER_01 1:06:34

Wow, thank you, thank you so much for describing it that way. And like I I've always believed these subjects that form the fundamental basis of everyone's education all over the world, they're there for a reason: math, physics, chemistry, biology, and so on. Everything else builds on top of them. Because even musical notes are fundamentally math equations. Um, AI, like you described, it's basically predicting the next word and the next sentence, all based on the weights given to words that have been learned by the machine learning that powers the AI.

SPEAKER_02 1:07:15

Yeah.

SPEAKER_01 1:07:16

So if one word weighs more than another word, then it's gonna pick that word, and then it's gonna pick the next word that weighs more, and so on, and so on. So it's something I also fundamentally believe in, and now I see statistics in everything. Yeah, you're more likely to notice a pattern because that likelihood falls on a particular Gossian bell curve, and now I'm talking like a nerd, I don't know what to tell you. It's you're exactly right though.

SPEAKER_03 1:07:42

You're speaking my language.

SPEAKER_01 1:07:46

Thank you so much for this insightful conversation. And it's it's just one hour gone, just like that. And I learned so much from you.

SPEAKER_02 1:07:54

Really fun. Yeah, thank you.

SPEAKER_01 1:07:56

Thank you, thank you so much. And of course, like I mentioned, it's gonna be ready in a few weeks, and the the profile picture that you want me to use, just you can send that over. And I was gonna ask, I mean, I could cut out all these final parts from the episode or leave it in. I was gonna ask, how did I do as a podcast host?

SPEAKER_03 1:08:22

I mean, it was really, really easy to discuss with you. I've been on a couple of podcasts, and sometimes, you know, the questions can be awkward or or anything. It's like a little scripted, you know, but it didn't feel scripted. It really felt like we were just having a cool conversation. And so, yeah, I really exactly, yeah. I really appreciate your curiosity and and your prep your preparation as well, right? Like you could really ask me some some questions that make it easy to dive deep quickly. So I really appreciate that. Yeah, it was great.

SPEAKER_00 1:08:56

Thank you so much.

SPEAKER_03 1:08:58

Thank you too, Vid.

SPEAKER_00 1:09:00

Have a good evening. Talk to you soon.

SPEAKER_03 1:09:02

Thank you. Talk with you soon. Uh, you'll s you'll have an email from me, okay? Uh yeah, I appreciate your time.

unknown 1:09:08

Thank you.

SPEAKER_00 1:09:08

Talk to you later.

SPEAKER_03 1:09:09

Bye bye.