Agnikul Cosmos, the space tech startup incubated at the Indian Institute of Technology Madras, will apply for a patent for its full-scale, second-stage engine Agnilet, which it claims is the world’s first single-piece 3D-printed rocket engine. This means it’s been made in one shot without the thousands of parts that are typically assembled in a conventional rocket engine.
Agnikul Cosmos has also developed its own mobile launchpad, which allows it to launch rockets from any authorised port. The startup is awaiting final approvals from the Indian Space Research Organisation to launch its rockets from Sriharikota. By developing its own mobile launch pad, the startup is in a better position to scale up its business in the future.
In an interview with Moneycontrol, cofounder and CEO Srinath Ravichandran, a Wall Street trader-turned-aerospace engineer, spoke about why India’s space technology ecosystem has an advantage over those of its international counterparts because of its practices and policies and said that there is an immense business opportunity if the challenges of delivering satellites are addressed.
The startup was founded in 2017 by Ravichandran and Moin SPM, an operations specialist. It has raised $15 million, including $11 million last year in a Series A round backed by investors such as venture capital company Mayfield India, and angel investors Anand Mahindra and Nithin Kamath. Edited excerpts:
What are the current challenges in the space tech industry?
In the last 20 years, with electronics shrinking, satellites have also shrunk in size. Quite a bit. But still, rockets, the vehicles that carry satellites to space, are still stuck in the past. And the reason is very straightforward. Satellites are mostly electronics. But rockets have a lot of other pieces – you have the tank, you have the engines, you have a lot of other moving pieces. So, what ended up happening is all those who manufactured small satellites had to share their ride to make it economically viable.
The problem with this rideshare model is that it is not very efficient. I'll use an example of people waiting at a bus stop. So, you'll have to wait for the right bus, it should be going to where you want to go, it should have space in it for you to get on board. All of these problems are there.
Additionally, since it’s space, these satellites are very costly. There is a certain level of exclusivity and not everyone wants to ride alongside everyone. They are afraid that if I let this satellite ride with me, mine may get damaged. So, people end up waiting and the wait time is pretty bad – almost along the lines of two years.
Where does Agnikul Cosmos come in?
We wanted a technology that, say, if someone made a satellite in two months, they should be able to launch in two weeks. This is when we started getting into every single detail on how to reduce waiting time. We looked into manufacturing, design, and architecture techniques that would help in developing a vehicle in two weeks. And despite all that, there was one problem that was lingering. And the problem was that whatever you did, there still was some level of fixed size that was going to be required. One can probably come down from a 1 tonne capacity vehicle to a 100 kg capacity vehicle. But what do you do when someone wants only 30 kg?
So that’s when we decided that beyond all of this manufacturing speed and designed feed and all that that we also do, we should have an inherent core architecture where the vehicle itself is scalable. We can increase or decrease the weight based on what the customer, the satellite owner, wants. And now we are able to say that we can almost offer a dedicated launch.
What makes Agnikul Cosmos different from others?
So, we do three things very differently from what, in my understanding, anyone else is doing. One is customizability for scaling up and down – usually people build different vehicles for different capacities. There is usually 100 kg, 500 kg vehicle, and 1,000 kg vehicle, in terms of carrying capacity.
But here, we did away with that and it’s just one vehicle that we scale up or down. It’s a smooth change between different payloads so that we are really not punishing the customer by forcing him/her into one or other buckets.
Second is the way in which we make our rocket engines – and this came from the fact that if you look at what takes the longest time to make in a rocket, it is the engine. A rocket engine has a lot of parts that have to be welded carefully. So what we want to do is to come up with an engine that can be made in an extremely short amount of time, within a week’s timeframe, so that it does not become a showstopper. That’s when we took up 3D printing, where instead of multiple parts, one can put it all together and do it as one part.
With that idea in mind, we set aside to build various iterations of rocket engines. And slowly we have gotten to a point – the team has done an amazing job here – where the entire rocket engine is 3D-printed in one shot. There is basically zero assemblies in the making of a rocket and no human intervention.
Thirdly, we have built our own mobile launcher, through which we can set up shop in any authorised launch port and launch from there.
What do you need from the Indian Space Research Organisation?
From a policy standpoint, basically, clearances. That’s the only real thing, other than launch port services. For clearances, we will have to give our hardware specifications very clearly because we will be bringing it into another facility. It has to be compatible with ISRO’s infrastructure.
How long does it take?
It depends on the exchange of information from both sides. I would think three to six months maybe, or maybe a little more than that because a lot of details have to be ironed out. The idea is always simple. But then if you look at it and start looking at all of the small details, it's like, okay, this is slightly off, that's slightly off. So we are also trying to be mindful of the fact that ISRO has its own plans.
How are you on funding and what’s next for Agnikul Cosmos?
We raised three rounds of funding, a total of $15 million. We raised it from mostly institutional investors and a few family offices have also participated like Anand Mahindra.
In the future, we want to scale up more. Right now, we have two-three launches a year. Getting to 25 in a year would be an important aspect. And we are also looking to make our vehicle more customisable. Even for a few kg, I should be able to build a slightly different rocket that will make the cost lower and also address the needs of the customer.
How do you zero in on talent for your company?
I think ISRO has done an amazing job with the technology part. I think we are among the best players in the world known for rocketry. I think there are so many good engineers who come out of college. And, in fact, that’s the way we operate. We talk to a lot of retired ISRO officials. And we have a very young team, where no one has actually built rockets, but they are all deadly serious and dead passionate about building rockets. And we were able to put these two together to get to the next level.
What cybersecurity policy do you employ to keep your customers’ data safe?
All the data handled by people is monitored. We have limits on what data can be downloaded onto laptops, we have limits on what can be uploaded, we have limits on who can access what information – access restrictions exist even within the team. We try to get 100 percent traceability of data. The data ocean is vast, but we try to do it.
How does the Indian space tech ecosystem fare compared to the rest of the world?
India has a capable ecosystem that has actually been developed by ISRO. We have vendors who understand what it means to build stuff for space, we have vendors who understand the rigor that ISRO imposes on them. And these things have been happening for decades already. I think that is actually an advantage of doing space tech in India. We are not, you know, freshers to space as a country. Yeah, there are some new technologies to obviously focus on, but ISRO is pushing the boundaries.So the knowledge exists, it’s spread out across scientists, vendors, advisors, professors, and they are all accessible. I think we just need to put in the effort to bring a good system around it so that we can derive this knowledge and try to add value in a business problem, and that’s what we are trying to do.