Meeting created at: 15th Jul, 2026 - 11:00 PM1
Elise de Reus: So rather than thinking of Cradle as a product that a gift that you bring home from the store and it has a bow around it and you unpack it at home and now you can use it.
Elise de Reus: It is a platform that you can learn how to interact with, you can collaborate with, you can plug it into your existing workflows, but you can also start to reimagine what those workflows look like, given that you now have machine learning at your fingertips.
Flo Lumsden: Welcome to Talk Bio to Me, the show where we sit down with the people building the future of biology.
Flo Lumsden: I'm your Flo Lumsden, founder of Chorus Studios, a podcast production company where we transform complex technologies and breakthrough ideas into human stories that educate, entertain and inspire.
Flo Lumsden: This year we headed to synbio Beta, the world's largest conference for synthetic biology.
Flo Lumsden: This mini pod features some of the most fascinating founders.
Person: We make spores that actually eat the insects.
Flo Lumsden: Scientists.
Person: We have carbon fixing microbes that can essentially eat emissions.
Flo Lumsden: Researchers.
Person 1: This is gonna DNA into a programmable material that we can use to turn biology into infrastructure investors.
Persons 2: Because it's garbage in, garbage out on AI.
Person 3: And Mother Nature still works in mysterious ways, right?
Flo Lumsden: And AI innovators.
Elise de Reus: I'm increasingly excited about AI actually making.
Flo Lumsden: An impact in the lab, shaping the future of biotech.
Flo Lumsden: From breakthrough therapies to AI that's beginning to design biology.
Flo Lumsden: I discovered one of the most exciting technological revolutions of our time.
Flo Lumsden: Okay, Talk Bio to Me.
Flo Lumsden: Hi, Elise.
Elise de Reus: Hey.
Flo Lumsden: Thank you so much for being here.
Flo Lumsden: I'm Flo from Quora Studios and this is Talk Bio to Me, a mini pod I'm creating just for synbio Beta and I'm excited to talk to you about Cradle today.
Flo Lumsden: Could you give a little introduction to yourself, just your name and your role?
Elise de Reus: Thank you for having this podcast and sharing the stories of symbiombio Beta.
Elise de Reus: My name is Cradle and I'm a co founder at Cradle, where I lead the customer success team.
Elise de Reus: So I work with all the fantastic customers across biopharma and industrial bio to help them get on their way with AI guided protein design on our software platform.
Flo Lumsden: Very cool.
Flo Lumsden: How long have you been working with Cradle?
Elise de Reus: We founded the company four and a half years ago and then we had a good amount of time brainstorming on what we thought the industry needed and what kind of business model and company would fit that best.
Flo Lumsden: Yeah, I was going to ask you, what problem is Cradle working to solve, but I was also going to ask you how the founding team came together.
Flo Lumsden: Maybe those things, those two questions feed into each other.
Elise de Reus: They do so my background is in bioengineering.
Elise de Reus: I'm a wet lab biologist and I was focused on high throughput experimentation at a company called Zymerton here in the Bay Area.
Elise de Reus: We were trying to scale biology and bring bio based products to market faster using robotics and automation.
Elise de Reus: And while at Zymogen, I realized that's part of the solution, but the other part of the solution really needs to be what designs are we putting into those experiments?
Elise de Reus: Our co founders, Eli and Steph were inside of Google working on teams like Machine Intelligence, Google Brain Accelerated Sciences.
Elise de Reus: And so they were seeing where machine learning models were already starting to impact language.
Elise de Reus: So language models, but also the natural sciences in protein space.
Elise de Reus: So we thought, why aren't our types of people collaborating yet?
Elise de Reus: And why is there no wet lab at Google and why is there no Google team at Zymergen?
Elise de Reus: And so how about we build the company that brings together machine learning innovations and biotechnology in a way that helps scientists design better proteins in less time and fewer experiments.
Flo Lumsden: Very cool.
Flo Lumsden: Yeah.
Flo Lumsden: I was hearing from Jared that it's like a feedback loop that you created to like fail more quickly, learn more quickly, create better protein designs.
Flo Lumsden: I don't, I'm new to the field.
Elise de Reus: You're saying it already.
Flo Lumsden: We'll have to correct my terminology, but it sounds really smart.
Flo Lumsden: Like it sounds like an obvious next step for the industry.
Elise de Reus: We think it's needed.
Elise de Reus: There are foundation models that can model all aspects of proteins, but they won't necessarily be experts at the particular hard, gritty problem that a scientist is trying to solve.
Elise de Reus: Whether they are trying to develop antibody that binds a cancer cell but not a healthy cell.
Elise de Reus: Whether they are developing an enzyme that has to be super effective, but also withstand the harsh conditions of your laundry machine.
Elise de Reus: You need to have machine learning models that understand what success means.
Elise de Reus: You have to have a way as a scientist to translate your goals into machine learning constraints, machine learning speak.
Elise de Reus: And you have to have a way to collaborate with this machine learning model.
Elise de Reus: You have to understand what it's telling you, where it thinks it's going to be really successful, when it's worth taking new praise into the lab, or when maybe you have to course correct.
Elise de Reus: And all of that is much more than just a simple model.
Elise de Reus: It is really a fully fledged product that helps you design proteins faster and more effectively.
Elise de Reus: And that's what we're really building at cradle.
Elise de Reus: So we enable scientists to create this loop, running their experiments, generating new protein variants with machine learning, and then informing these models with new data coming back from the wet lab.
Flo Lumsden: Amazing.
Flo Lumsden: Very, very cool.
Flo Lumsden: You were mentioning you work in pharmaceuticals and bioindustrial applications, and you've done several large projects with large pharmaceutical and bioindustrial companies.
Flo Lumsden: Was there anything that surprised you as you started to deploying that platform on real projects?
Elise de Reus: It's.
Elise de Reus: We're.
Elise de Reus: One of the most interesting and exciting parts of building Cradle is that we are building a product that is in the hands of scientists around the world in leading companies, whether they are biopharmaceutical companies or biotech companies, in the chemical space or the ag space.
Elise de Reus: It's really where the rubber hits the road.
Flo Lumsden: Right.
Elise de Reus: What has been surprising is how much better the outcomes of a project are when there is this collaborative approach from the start.
Elise de Reus: So rather than thinking of Cradle as a product that a gift that you bring home from the store and it has a bow around it and you unpack it at home, and now you can use it.
Elise de Reus: It is a platform that you can learn how to interact with, you can collaborate with, you can plug it into your existing workflows, but you can also start to reimagine what those workflows look like, given that you now have machine learning at your fingertips.
Elise de Reus: So that collaborative process and designing workflows that are just more ML native from the start has been surprising and really rewarding.
Flo Lumsden: Very cool.
Flo Lumsden: Can I ask you what ML Native means?
Elise de Reus: ML native, Absolutely.
Flo Lumsden: So machine learning.
Elise de Reus: Machine learning.
Elise de Reus: Okay, got it.
Elise de Reus: So rather than processes that are existing and you have machine learning as an afterthought.
Flo Lumsden: Right.
Elise de Reus: You really integrate machine learning into the workflow.
Flo Lumsden: Right.
Elise de Reus: That's what AI scientists call lab in the loop.
Elise de Reus: But as a lab scientist, I call it AI in my wet lab loop.
Flo Lumsden: Right, yeah.
Elise de Reus: The next step is to reimagine that workflow, reimagine that process and go.
Elise de Reus: Now that we have machine learning capabilities, how do we produce the types of data sets that really inform these models in the optimal way?
Elise de Reus: So how do we work with this new tool?
Elise de Reus: How do we work with this new tool?
Flo Lumsden: Yeah, it's like you discovered scissors.
Flo Lumsden: What am I going to do with the scissors?
Flo Lumsden: I'm going to cut things, you know, and I'm going to figure out how I can do my job better with this new tool.
Elise de Reus: Yeah.
Elise de Reus: Or what?
Elise de Reus: Maybe there are different questions that you can ask or different craft projects that you can make now that you have scissors.
Elise de Reus: Yeah.
Elise de Reus: Yes.
Flo Lumsden: Different shapes.
Flo Lumsden: Very cool.
Flo Lumsden: Thanks for going along with that analogy.
Flo Lumsden: So in terms of human health and sustainability, why does this innovative platform matter to the average person?
Elise de Reus: When you, as an average person interact with A designed protein a drug, in a medicine or in your laundry detergents or indirectly because the food that you eat has benefited from better fertilizers or better crop protection.
Flo Lumsden: Right.
Elise de Reus: Many steps have been taken to get that product into your hands.
Elise de Reus: So there's a whole regulatory, there's scale up, there's commercial.
Elise de Reus: But in the developing the fundamental protein, that's where there is a huge opportunity for platforms like Elise_Cradle or machine learning approaches to innovate faster.
Elise de Reus: So to get better molecules out to the market faster, but also with new types of properties.
Elise de Reus: So you can, yeah, you can develop fundamentally products that didn't exist or couldn't have existed in like a decade earlier because maybe there you'd have to make specific trade offs in activity of an enzyme or its stability or its ability to withstand your laundry detergents.
Elise de Reus: As now you have a machine learning model that can help you as a scientist really find the solution that ticks all those boxes.
Flo Lumsden: Very cool.
Flo Lumsden: I was hearing that in the antibody protein space, you're working in that space or you have clients that work in that space and that there could be just a speeding up of discovery of solutions or treatments for cancer and autoimmune issues.
Flo Lumsden: Do you have any examples of that or any stories you could share?
Elise de Reus: So the regulatory process, the clinical trials and everything, that takes many years.
Elise de Reus: The technologies that we're seeing applied now in the protein design or antibody design space aren't that old.
Elise de Reus: Like when we started Cradle four and a half years ago, people still were very skeptical of whether AI would ever be practically useful in this type of research.
Elise de Reus: So as you can imagine, it's going to take a little bit of time before we can really point at something that passed through clinical trials and say that was AI designed.
Elise de Reus: What we are already seeing is that there's all sorts of things where AI or machine learning can help, like maybe some of the properties of antibody you don't have to measure in the wet lab because you can predict behavior ahead of time.
Elise de Reus: Maybe you don't need to go through immunizing a mouse or a llama to get initial binders that you can turn into your antibody, but you can design these de novo.
Elise de Reus: All of these are advances that we're seeing now in the protein design field.
Flo Lumsden: So it may help speed up the process just a little bit.
Flo Lumsden: Getting the drugs, it can speed up.
Elise de Reus: The process a lot.
Flo Lumsden: Okay.
Elise de Reus: Yes.
Flo Lumsden: Okay.
Elise de Reus: Yeah.
Flo Lumsden: Cool.
Flo Lumsden: I have an autoimmune issue, so I'm waiting for the next drug.
Elise de Reus: Okay, we're on it.
Flo Lumsden: I'm Excited about innovations in treatments for rare diseases and things like that.
Flo Lumsden: So that's how it matters to me,.
Elise de Reus: Is working on it.
Elise de Reus: And are they here at the conference?
Flo Lumsden: It's Ehlers Danlos syndrome.
Flo Lumsden: I'm not private about it.
Flo Lumsden: I don't know.
Flo Lumsden: I don't know.
Flo Lumsden: I don't know.
Flo Lumsden: I haven't heard anybody talking about Ehlers Danlos syndrome, but I might.
Flo Lumsden: I'm gonna go to the femtech event after this and you know, a lot of women have Ehlers Danlos syndrome, so maybe they'll be talking about it in there.
Elise de Reus: Super interesting.
Elise de Reus: I'll have to go and learn more about it.
Elise de Reus: Yeah.
Flo Lumsden: So why did you decide and your company decide to come participate in such a big way at Synbio Beta this year?
Elise de Reus: We think Symbio Beta is a unique conference.
Elise de Reus: It has gotten more and more diverse every year since I started attending four years ago.
Elise de Reus: I see the share of different types of scientists.
Elise de Reus: Fursona interests really expand.
Elise de Reus: So there are a lot more biopharma attendees over the past few years like Oracle and Nvidia all of a sudden joins and takes an interest as compute and models are becoming more important in this space.
Elise de Reus: And the commonality between all the participants at Symbio Beta is that I think they're all very motivated by biology, very hopeful for what it can do for the future of the world, and may be willing to take some risks here and there.
Elise de Reus: So think creatively and learn from each other in this like cross pollinating way.
Elise de Reus: And I think that's quite unique to Symbio Beta as a meeting.
Flo Lumsden: I've heard that from others that it's a unique mix of different parts of tech and biotech and industry.
Flo Lumsden: Is there anything else?
Flo Lumsden: Do you have any updates from Cradle?
Elise de Reus: It wasn't so long ago that we put out our white paper where we describe cradle one, the platform, the 2026 platform.
Flo Lumsden: Okay.
Elise de Reus: I encourage listeners to go check it out if they haven't already.
Flo Lumsden: Perfect.
Flo Lumsden: We'll put it in the show notes.
Elise de Reus: Cool.
Flo Lumsden: Thank you so much, Elise.
Elise de Reus: Thank you for having me.
Flo Lumsden: Yeah.
Flo Lumsden: Thank you for coming.
Elise de Reus: Yeah.
Flo Lumsden: Don't forget to subscribe or follow talk via to me on Spotify, Apple Podcasts, YouTube today for episodes dropping very soon.