Loading summary
Transcript20 lines
- [00:02]
A
Hi everyone.
- [00:03]
B
I'm Rowena and you're listening to the Africa Health Ventures podcast. In this podcast, we're going to unpack what's going on in healthcare in Africa today. But that's not all. We're also looking at the innovations, the ventures, the entrepreneurs that are going to leapfrog ahead in the next 10 years. Now, I admit we've been a little quiet lately.
- [00:27]
A
What can I say?
- [00:28]
B
We launched a venture capital fund this year and turns out it's a lot of work. Yeah, but it's been worth the wait. We've got a great episode for you today. For those of you investing in Africa,
- [00:41]
A
by which I mean those of you
- [00:43]
B
that are seeking financial returns, the question you're always asking yourself is where are the exits? Regardless of whether you're investing in healthcare or not, the answer that everyone is talking about is instadeep. Instadeap was not only the biggest healthcare exit on the African continent, last year
- [01:05]
A
it was the biggest exit of any
- [01:07]
B
company in any sector in Africa. Instateep was acquired for 500 million euro. Not too shabby. The purchaser was a company called Biontech, which you know as the first company to develop a COVID 19 vaccine. This is remarkable for a couple of different reasons. One, it's a healthcare success story. At a time when the vast majority of investments in Africa are in fintech financial technology, it's also a deep tech play. Instadeep is an AI artificial intelligence company known by the geeks for its multi agent reinforcement learning. And let's be honest, this all comes at a time when most of the world is skeptical about whether deep tech can even be built. And can it be scaled in Africa? So how did a math geek from Tunisia build a company of a few hundred people that captured the hearts and minds of a continent? Let's find out before we dive in. A quick announcement. If you're interested in healthcare ventures in Africa, subscribe to our newsletter. To get all the latest going ons, you can find it@africahealthventures.com News and if you want to nominate a promising seed stage startup bringing better healthcare to Africa. Or if you want to learn more about investing in African healthcare, get in touch with us through our website@africahealthventures.com One last friendly reminder. The content here is for informational purposes only and should not be taken as legal, business, tax or investment advice or be used to evaluate any investment or security. And now back to Karim Bagheer, the founder of instadeep. My first question For Karim was Africa needs more exits and you are the success story of 2023. For all the founders listening to this podcast, for the ones that have been, you can't do deep tech in Africa, you can't do health tech. What do you have to say?
- [03:17]
C
I would say the most important message is that it is possible. That's important because when we started, and this was 2014, literally 10 years ago, it was not obvious that such a bet would be successful. And actually many of my friends, ex colleagues said that we were just dreamers and this would never happen and everything. That's the first message. It's possible. Across those 10 years, we learned a lot. Instadeep started as a passion project. We started literally as a bootstrap. $2,000, two laptops in North Africa, in Tunis, and built up the company from there. And perhaps among the few takeaways from that journey, I'll share a few. I think the most important thing is to be genuinely passionate about what you are building because it is a very long road, a very hard path. And if you do not have the passion on top of the competence to push the project forward, well, you're going to get at some point, like you're going to want to give up or something. And so for me, this was actually the whole goal of the project was to prove that young Africans could innovate, that we could be doing amazing things in deep tech. In AI, there was the example of DeepMind at the time that had just gotten acquired by Google, which was very inspiring to us. And when we were telling people, hey, we're going to do AI innovation like DeepMind does, people were sort of like in disbelief. And yet it happened many years later. Actually Google led by Google AI invested in Instady. And then we had indeed the exit last year with BioNTech Group. And so really the message is be passionate about what you do. That is extremely important. If you have the mix of passion and competence, then things become possible. Another important lesson that I'd like to share is in the early days we were not sure about the business model. Like should we be covering the global market or should we be focused on the local market? And my advice to founders in Africa is have international ambitions from day one. It doesn't mean that you should not be covering the country where you are based. You can absolutely do that, but do not limit your horizons. Very often I see a failure of ambition, people building locally, while you should be international and have global ambitions from day one. And this is going to maximize your chances to generate revenue and so on. Particularly if you are building in a digital manner, which I believe is also easier in Africa. I think there's a huge difference between the prospects of a hardware startup in Africa and a software or digital or AI startup in Africa. And so my advice is try to focus more on the digital side of things because it scales better and it allows you to get international markets faster.
- [06:06]
A
Amazing. Karim, thank you. And I will encourage our listeners to check out the Rest of World article that you did. It's fun to see some of the early experiments that you did and how far it was from Insta Deep and where it is today. Can I probe a little bit deeper just on that exit question? And again, having spoken with so many founders that have a great idea, they have initial traction, maybe they're at series A. What are some of the things that you did right on your path to exit to demonstrating that gold standard? Was it about fostering relationships or were there some strategic decisions that you can point to that greased the wheels in terms of you demonstrating this successful exit?
- [06:44]
C
I think the, the story of NIP really like there was always a sort of compromise between different pressure points, for example, like how much to build on the continent in terms of teams, offices and outside the continent, in this case Europe and now even the US for instance. And we found a good compromise where we had a slight majority of the team in Africa and growing offices also in Europe. And this created a sort of positive emulation and a culture which was both African founded but also international. And that was a good balance. Same way in terms of balance between innovating, which is like algorithmic innovation, fundamental AI research innovation and the like, and practical applications. I think this is one of the aspects we got right and which is a challenge always for founders. There is value in fundamental AI research. That was one of the surprising things, at least for us. When we published our first paper, we were not sure if there was value into publishing the paper, for example, versus patenting the technology, et cetera. And it turned out to be even from a business standpoint, let alone open source and good karma point of view. But even from a business standpoint turned out to be the right decision because it put us on the map and so that was useful. At the same time, you want to be very pragmatic. I mean, we were a bootstrap. So there was always this tension about generating revenues. And it's almost like you have to see the initiatives you take as a founder of a startup, as there are some initiatives that are short maturities like generating revenue today and perhaps it's not the best project you could have. You may you have to make some trade offs but still it's revenue flowing into your company and you have longer term bets, sometimes very long term. There's. I'll give you an example. There's one product, DPCB AI which is today a leader in printed secret board design powered by AI, full autonomous. It took us more than five years Rowena to build and refine and now it's the best in the world. But that was a very long dated project. So you want to have some long term bets and also like some short term, hey, let's feed the team, let's get revenues, let's keep the momentum. And you have to have that balance and usually investors will like that balance. They like to see you have some short term revenues but also the potential to do much more ambitious thing. And I think like to answer your question about the acquisition, I think this was the right mix because when we started working with Biontech Group like we did with the other partners we have, we always had this culture of delivering value to our partners. I had almost this motto that if somebody is going to invest money with us, they need to get at least a 10x return on the investment they make with Instadeep and committing to working with us or like throwing us revenues, it has to work for them. And it is this win win mentality that actually led to further developments with Biontech Group like we did with several other customers when we did our 100 million series B actually we had several partners invest obviously Biontech, Google investors as well. Deutsche Ban, the national railway company of Germany invested as well. So for me this was a validation that putting the partner first, making sure they get value is key because this will ultimately generate more business, generate more opportunities and like it happened to us, actually leading to an exit.
- [10:11]
A
That is fascinating Karim, and that is so useful to see how you position yourself particularly around these international partnerships that create value on a day to day basis and obviously created value when you did your exit. I also really appreciate your point about open source because it has a complicated relationship on this continent. I think many founders are averse to open source because there's so much emphasis on cash today. But you've made a really important point that that can be a really powerful marketing and partnership building activity in addition to contributing to core science and moving the ecosystem forward. I think it's really valuable to hear from you. In particular, I want to go back to the first point that you mentioned just about your team and how it's composed. I was speaking with an investor last week who literally said there is a shortage of entrepreneurs, there's a shortage of talent in Africa. I would love to hear what your response is to that statement. Particularly as someone who decided to place the majority of your team in Africa. What is your reaction to the question of investable talent on this continent, particularly in the deep tech or the biotech
- [11:15]
C
space or investable projects? And I actually agree that today this is the limiting factor. If you look at, for example, funding, things were very different ten years ago. Ten years ago it was actually very hard to get a project funded, especially like early stage. Today this is very different. There's a lot of money, international VCs or local ones that are chasing projects and it's more like the right kind of projects that are missing. So this is a very interesting question actually, because it's really like the chicken and egg situation. If you, for example, like want to build an amazing deep deck or amazing AI company, well, you need top AI talents or top deep tech talents, but they will only come to you if you are already a great one. And so there's this chicken and egg situation which ultimately in all case, we resolved by training ourselves. My background is applied mathematics. I trained neural nets back in the states more than 20 years ago. And so when I read the paper about AI, this was in the very early days of instageep and I realized, hey, actually I know this thing. I could be good again at these things. Like, this is not science fiction to me. Is this really what they're talking about when they're talking about deep learning? And I was like, I can do this. And so effectively I rebuilt myself into an expert. And this is an advice I'd like to give to founders of a deep tech company. You have to be technically competent, you have to lead from an innovation, from a deep tech innovation point of view, your company. And so if you're not experts in the domain, become an expert or focus on a domain where you can be a world class expert. You have to be among the best in the world at what you do personally, because only then can you sort of like show the way forward and educate and bring up young talents. I mean, there are formidable talents in Africa, but what you need is to create the right environment and sort of bring the mentoring. And so if you are yourself an expert in the domain, it becomes a lot easier. And this is exactly what I did in the early days. We couldn't afford to hire like Top notch talents in a very competitive market, whether it's in Africa or in Europe or the us. So I literally rebuilt myself into an AI expert and then started hiring young talents out of university. They became amazing, they became world class. Now they lead some of our teams. I want to mention, for example, Alex Letter, who joined us right after university and now leads a team of like 50 top notch AI researchers. And there are many other examples in the company. And so that's how you solve this sort of like talent qualification challenge is by becoming yourself as a founder, world class into what you do. And you cannot be the founder of a deep tech company or an AI company if you are not yourself world class and technical. And so once you pass that, then sort of like bring the whole team up to speed up to the desired standards, but you set the bar and you have to set it high. Now the good news is this actually can work. And in our case it worked both in Africa and also in Europe, where same kind of approach proved out to be fruitful. I had studied at Ecole Polytechnique in France, which was the top engineering school of France, and so I managed to get a few interns and then they loved working with me and then they stayed and so on. This is driven by the founders, but to your point, I think this is the limiting factor indeed. Today on the continent, we need more world class founders. And once we have those world class founders, everything else will come relatively quickly, funding opportunities, et cetera.
- [14:49]
A
That is a very clear roadmap. You start with the idea, then you work on yourself. You make sure you're bringing the leadership that you need to bring. You create the environment and then the talent will come find you, be it in Africa or in Europe. Thank you for sharing that roadmap, Corinne. The last question I have for you, because this is the Africa Health Ventures podcast, is to talk a little bit about the opportunity that you see, particularly in healthcare and in biotech. What do you see in terms of the future, in terms of opportunities for AI to solve the world's most pressing healthcare challenges?
- [15:22]
C
I think we are really at an extraordinary moment. If you look at the progress in AI, we've had amazing progress in terms of natural language processing with the chatgpts of the world, et cetera. And so if we were to compare where we are today in biology, biopharma, healthcare, I would say we're probably like it was in 2020 when GPT3 came out. So GPT3, it kind of worked. It was quite remarkable, this notion that you could do zero Shot, question and answers, or a few shots, putting it in the context window and in the prompt, et cetera. But it was not perfect yet. And what we've seen since then over the last four years is an incredible exponential progress in the capabilities of those systems. I actually believe we will see exactly the same story play out in biotech and healthcare over the next four or five years. So we already have very powerful models, some of them actually developed by the teams at Instadip and Biontech in genomics and other fields. But really the best is still to come, and in particular on the application layer. If you look actually on where we are in the field of biopharma, I would say, and this is in particular for the continent, we still have limitations in data. So what is holding model innovation, product innovation in the field is actually lack of relevant data. So for all the African founders that are listening to us, there is actually an opportunity to build up the data assets of the future. Think about like whole genome sequencing, for example. Now you can get a whole genome sequence for almost $100, like between $100. I'd love to see many more people working on building those data assets in Africa for Africans, because this will drive the next generation of personalized therapies, personalized diagnostics and the like. And so ultimately, AI is a game of data. Of course you have model innovation, you have compute scaling, but really the limiting factor today is data. And so from that perspective, I think we have great opportunities ahead of us and for sure we can improve the standards compared to where we are today. So I'm extremely optimistic that over the four, five years, like we've seen in nlp, we'll see really incredible innovation led by more data, more model innovation, more compute. That golden age is only starting. And this creates a huge opportunity for founders.
- [17:55]
A
Phenomenal. Phenomenal. And just to echo the opportunity that you're talking about, Africa is the most diverse continent on the world. The wealth of genomics data that exists on this continent, that can be applied to identify novel therapeutics, identify disease pathways, is huge. And we're just at the tip of the iceberg in terms of what's possible. Is there anything else you'd like to add for founders, investors and other ecosystem builders, or should we wrap it up there, perhaps?
- [18:20]
C
I'll just add that this is an extraordinary time. The AI opportunity is actually bigger than anything you think. Some people think it's overhyped. I actually think it's underestimated. And so my advice to founders who are listening to us is do not limit yourself. When I started Instadeep, we did it for the right reasons, but we never thought actually this could work to the scale at which it did. And this was really a discovery. So my advice is really set the bar high, go for it, do what you love and don't take no for an answer. You can build world class projects in Africa to the world, and this is an extraordinary good time to do it.
- [19:02]
B
And that's a wrap. That was our conversation with Karim Baghier, the founder of instadeep. Let me recap a few of his points, much less eloquently than he did. First, start with deep intrinsic passion. Then build the idea, the project and the talent, in that order. When you bring on clients or investments, make sure you're returning exponential value. And perhaps my biggest takeaway we live in an extraordinary time. The best is yet to come.
- [19:40]
A
If you like what you heard today,
- [19:41]
B
don't forget to subscribe to this podcast. We also love positive reviews. Please. And if you want to stay in the know about what's going on with healthcare ventures in Africa, you can also subscribe to our newsletter@africahealthventures.com News thank you for tuning in to our show today. We'll see you next time. Bye bye.