CoreWeave Expands AI Cloud to India with 240MW Navi Mumbai Campus, Prioritizing High-Density Cooling for Sustainable AI Growth

CoreWeave has announced its entry into the Indian market through a partnership with AdaniConneX, establishing a 240MW AI data center campus in Navi Mumbai. This new campus will feature three 80MW buildings at AdaniConneX's Taloja site, with CoreWeave as the sole tenant and an option to double capacity to 480MW. A key technical highlight is the integration of direct liquid cooling (DLC) within the GPU data halls and a non-evaporative chilled water system, specifically designed to handle high-density AI workloads. The first phase of this facility is expected to be operational by mid-2028, deploying NVIDIA's Vera Rubin Platform. This development is significant for practitioners in cloud and DevOps as it directly addresses the escalating energy and cooling challenges posed by advanced AI. As AI models become more complex and ubiquitous, the power density within server racks dramatically increases, pushing traditional air-cooling methods to their limits. CoreWeave's commitment to DLC and non-evaporative cooling solutions demonstrates a practical approach to maintaining performance while attempting to manage the environmental footprint. This is particularly relevant in regions like India, where sustainable infrastructure development is a growing priority alongside rapid technological adoption. This expansion fits within a broader industry trend where hyperscalers and specialized AI cloud providers are increasingly investing in purpose-built infrastructure. The demand for AI compute is driving unprecedented data center growth, with projections indicating a significant rise in electricity consumption globally. Companies like Microsoft, Amazon, and Google are all grappling with the energy demands of AI, leading to innovations in data center design, power sourcing, and cooling technologies. The shift towards liquid cooling is a direct response to the heat generated by modern GPUs, which can exceed 150 kilowatts per rack, making efficient heat removal critical for operational stability and energy efficiency. In practice, this means that organizations planning large-scale AI deployments should prioritize evaluating cloud providers not just on compute capacity, but also on their underlying infrastructure's sustainability and cooling capabilities. The trade-off often involves higher initial investment in advanced cooling systems versus long-term operational cost savings and reduced environmental impact. Practitioners should watch for further innovations in thermal management, energy reuse (e.g., waste heat recovery), and the integration of renewable energy sources directly into data center operations. This move by CoreWeave signals that specialized, energy-efficient infrastructure is becoming a competitive differentiator in the AI cloud landscape, and understanding these nuances will be crucial for making informed architectural decisions.