Pune: Agnikul Cosmos, a private end-to-end space transportation company, has announced its capability to host Orbital AI Data centers for customers using its existing extendable upper-stage platform.
In its first commercial agreement of this kind, Agnikul will launch an orbit-based AI data center module developed by NeevCloud, India’s deep-tech AI SuperCloud platform, into space.
NeevCloud will deploy its space-based AI data center and run real-time AI inference applications on this patented platform designed and developed by Agnikul.
Also Read: Agnikul Cosmos Raises USD 17 Million to Boost Reusable Rocket Development
Orbital AI Data Centers: Platform to Enable Real-Time AI Inferencing in Low Earth Orbit
NeevCloud is among the first enterprises testing AI inference and cloud workloads on Agnikul’s patented extendable upper-stage platforms in Low Earth Orbit (LEO).
This collaboration opens a new revenue stream for Agnikul through launch and on-orbit services, removing the need for customers to design or deploy satellites.
The approach addresses latency limitations of terrestrial data centers by enabling real-time AI inferencing across regions underserved by ground-based infrastructure.
Agnikul is targeting a fleet of orbital assets equipped with AI stacks forming modular Space Data Center Modules (SDCMs).
The alliance serves as a key enabler for NeevCloud’s AI SuperCloud scaling roadmap, allowing expansion to more than 600 orbital edge data centers over the next three years.
The constellation will provide continuous, real-time AI inferencing capability by offloading heavy compute workloads from terrestrial devices to AI chips in space.
“As with all rockets, with more launches, we naturally leave behind upper stages in orbit,” said Srinath Ravichandran, Co-founder and CEO of Agnikul Cosmos.
“Our convertible upper-stage technology lets these stages stay active and functional, turning them into usable assets that can host hardware and software in space including compute or data capabilities. That’s the next step for a space transportation company – you build, launch, recover, and then extend into orbit.”
“We are not just building a data center in space, we are building an entirely new layer of orbital inferencing infrastructure,” said Narendra Sen, Founder & CEO of NeevCloud.
“To truly democratize AI, we must decouple it from terrestrial limitations. By partnering with Agnikul, we are taking our AI SuperCloud to the ultimate orbit edge – Space. This partnership ensures access to the AI infrastructure that powers the next decade of global intelligence, low-latency AI not just to our cities, but to every village and border post, AI for all.”
“This collaboration validates our assessment that enterprises are ready for a more commercially viable way to access space infrastructure,” said Moin SPM, Co-founder and COO of Agnikul Cosmos.
“Lower upfront costs for launch and platform, plus flexible rental or data services, create predictable scaling economics across the full space lifecycle.”
Orbital AI Data Centers: Compute Hosting Platforms
While Agnikul’s rockets – Agnibaan – feature boosters capable of returning to Earth, the company is leveraging its patented convertible upper-stage, originally designed as part of its launch architecture, as a long-lived orbital platform capable of hosting AI and data compute workloads.
Over the past few years, Agnikul has transitioned from technology validation to preparing for repeatable commercial launch operations.
The company has designed and flown key rocket systems while establishing mobile launch infrastructure, mission control, and ground stations developed entirely in-house.
These capabilities now support reliable, high-frequency launch missions. Building on this operational foundation, Agnikul is extending its vertically integrated launch stack to include orbital compute hosting platforms.
Orbital AI Data Centers: Pilot Launch of 2026 End
By combining NeevCloud’s deep-tech expertise and sovereign AI architecture with Agnikul’s orbital launch capabilities, the collaboration aims to bridge the global AI access gap.
The initiative will place high-performance AI inference nodes directly into Low Earth Orbit (LEO), enabling secure, ultra-low-latency intelligence for billions of users and critical industries worldwide.
The first pilot launch of Orbital AI data centers is scheduled before the end of this year. Following validation, NeevCloud plans to scale the network to more than 600 orbital data centers over the next three years.
Orbital AI Data Centers: Strategic Impact & Market Leadership
By aligning their expertise, both entities aim to streamline the path from computational design to physical launch.
The collaboration positions India at the intersection of two rapidly growing markets – the $255 billion global AI inference market, growing at a CAGR of 19.2%, and the emerging orbital data center and space-compute economy.
- Sovereignty & Security: Critical AI workloads for defense and finance can be processed in orbit under Indian sovereign control, mitigating geopolitical data risks.
- Cost Efficiency: By utilizing Agnikul Agnibaan’s reusable upper stages, the partnership aims to reduce deployment costs and lower capital expenditure for orbital infrastructure.
- Sustainability: Extending spent rocket upper stages reduces potential space debris while maximizing launch utility. Solar power can also support energy requirements for orbital data centers.
