coreweave, inc.

overview

CoreWeave is a specialized cloud computing infrastructure provider focused on GPU-accelerated workloads for artificial intelligence, machine learning, and high-performance computing. The company operates as ‘The AI Hyperscaler,’ providing access to massive GPU capacity including over 250,000 NVIDIA GPUs across 28+ data centers globally. CoreWeave evolved from a cryptocurrency mining operation in 2017 to become one of the leading alternative cloud providers challenging traditional hyperscalers.

business model

CoreWeave operates a Kubernetes-native cloud platform built on bare-metal infrastructure, offering GPU compute, storage, and networking services specifically optimized for AI workloads. The company generates revenue through cloud service contracts with major technology companies, AI labs, and enterprises, with Microsoft representing 62% of 2024 revenue. CoreWeave’s model emphasizes cost efficiency (up to 80% cheaper than traditional clouds), rapid deployment (35x faster instance spin-up), and priority access to latest NVIDIA GPUs through strategic partnerships.

data center profile

global footprint

us portfolio (from database)

projects by state

specialization

primary focus: ai-ml, cloud, hyperscale

key differentiators:

• Specialized GPU-first cloud infrastructure with 250,000+ NVIDIA GPUs

• 80% cost advantage over traditional cloud providers

• 35x faster instance spin-up times through Kubernetes-native architecture

• Priority access to latest NVIDIA GPUs (H100, H200, GB200)

• Bare-metal Kubernetes with DPU-based architecture

financial highlights

strategy

corporate strategy

CoreWeave positions itself as ‘The AI Hyperscaler’ challenging traditional cloud providers by offering specialized, cost-effective GPU infrastructure for AI workloads. The company’s strategy focuses on vertical integration through infrastructure ownership (via Core Scientific acquisition), strategic partnerships with colocation providers, and maintaining technological leadership through preferential NVIDIA relationships. CoreWeave is transitioning from pure-play cloud operator to infrastructure owner to reduce $10B in future lease obligations and gain 1.3 GW of owned power capacity.

growth strategy

Aggressive expansion through multiple channels: (1) Geographic expansion from 28 data centers (end of 2024) to 38 by end of 2025, including European market entry (UK, Norway, Sweden, Spain); (2) Vertical integration via $9B Core Scientific acquisition to own infrastructure vs lease; (3) Strategic partnerships with Flexential, Digital Realty, Chirisa Technology Parks for rapid capacity additions; (4) Customer diversification to reduce Microsoft dependency (62$6B Lancaster project) and other strategic locations.

power strategy

Multi-faceted power strategy combining on-site generation, utility partnerships, and renewable energy commitments to support massive AI compute requirements

renewable commitment: 100% renewable energy commitment for European data centers (Norway, Sweden, Spain). Investing $3.5B in renewable-powered facilities. Pennsylvania facilities exploring mix of grid power, natural gas, and potential nuclear options.

nuclear partnerships:

• Exploring nuclear power options for AI infrastructure

• 1.1 GW of nuclear power secured for AI operations (reported)

major commitments

partnerships

power providers

technology partners

NVIDIA (gpu-compute) : Elite Cloud Service Provider in NVIDIA Partner Network. Priority access to H100, H200, and GB200 Blackwell GPUs. Fleet of 250,000+ NVIDIA GPUs. NVIDIA Quantum-2 InfiniBand networking. Strategic investment of $100M from NVIDIA in April 2023.

Kubernetes (orchestration) : Kubernetes-native architecture running on bare metal without hypervisor layer. Custom CKS (CoreWeave Kubernetes Service) with DPU-based architecture using NVIDIA BlueField DPUs

NVIDIA BlueField (infrastructure) : DPU-based architecture for offloading node and resource management, enabling advanced security and network policies

financial partnerships

leadership

Michael Intrator

Chief Executive Officer, Chairman of the Board, Co-founder

Co-founder and CEO of Hudson Ridge Asset Management, a natural gas hedge fund. Prior to that, Principal Portfolio Manager at Natsource Asset Management LLC managing over $1.2B in assets. Trading background in energy futures and environmental markets. Recognized vision for AI infrastructure needs five years before ChatGPT launch.

Led CoreWeave’s transformation from crypto mining to AI cloud infrastructure. Net worth reached $10B+ post-IPO. EY Entrepreneur of the Year 2024 winner in Technology Infrastructure category.

Brian Venturo

Chief Strategy Officer, Board Member, Co-founder

Co-founder with initial role as CTO for over 6 years before transitioning to CSO. Led development of cutting-edge software powering CoreWeave’s competitive AI cloud platform. Background in commodities trading.

Instrumental in CoreWeave’s technical architecture and strategic direction. Board member since April 2019. Net worth $4.7B+ as of 2025.

Brannin McBee

Chief Development Officer, Co-founder

Commodities trader focused on North American Natural Gas, Power, and Agriculture markets. Previous roles at Active Power Investments, Fourth Floor Coastal LLC (oil & gas E&P), and Windy Bay Power LLC as proprietary trader.

One of four co-founders. Transitioned from CSO to CDO in March 2024. Net worth $4.7B as of 2025.

competitive position

CoreWeave is positioned as the leading specialized AI cloud infrastructure provider, competing against traditional hyperscalers (AWS, Azure, Google Cloud) and other GPU cloud providers (Lambda Labs, Cerebras, Modal). The company established itself as the #1 alternative cloud for AI workloads through early NVIDIA GPU investments, specialized Kubernetes architecture, and strategic customer relationships with Microsoft and OpenAI. CoreWeave’s March 2025 IPO as the first pure-play AI infrastructure company validated its market position and disruptive potential.

strengths

• Largest specialized GPU fleet - 250,000+ NVIDIA GPUs vs competitors

• 80% cost advantage over traditional hyperscalers for GPU workloads

• 35x faster instance deployment through Kubernetes-native bare-metal architecture

• Strategic NVIDIA partnership providing priority access to latest GPUs (H100, H200, GB200)

• Explosive revenue growth - 737% YoY in 2024, demonstrating strong market demand

opportunities

• Massive AI infrastructure demand growth - training and inference workloads expanding exponentially

• OpenAI relationship expansion beyond current $22.4B commitments as o1 and future models scale

• European data center expansion into markets with data sovereignty requirements

• Core Scientific integration unlocks 1.3 GW owned power capacity + 1+ GW expansion potential

• Customer diversification reducing Microsoft concentration below 50%

threats

• Hyperscaler competition as AWS, Azure, Google Cloud expand GPU offerings and reduce prices

• NVIDIA developing direct cloud services potentially disintermediating CoreWeave

• OpenAI and Microsoft building proprietary AI chips reducing GPU dependence

• GPU supply constraints from NVIDIA limiting growth or creating allocation challenges

• Economic downturn reducing enterprise AI spending and contract values

projects

sources

• CoreWeave - The AI Hyperscaler for GPU Cloud Computing

• CoreWeave (2025-10-14)

• CoreWeave Investor Relations

• CoreWeave (2025-10-14)

• CoreWeave - Wikipedia

• Wikipedia (2025-10-14)

• Nvidia-backed GPU cloud provider CoreWeave surges to $19 billion valuation

• CNBC (2024-05-01)

• AI infrastructure startup CoreWeave raises $7.5 billion in debt deal led by Blackstone

• CNBC (2024-05-17)

• CoreWeave Secures $7.5 Billion Debt Financing Facility led by Blackstone and Magnetar

• Blackstone (2024-05-17)

• CoreWeave Secures $7.5 Billion Debt Financing Facility

• CoreWeave Investor Relations (2024-05-17)

• CoreWeave Announces $650 Million Credit Facility to Support Ongoing Growth

• PR Newswire (2024-10)

• AI cloud firm CoreWeave raises $7.5 billion in debt financing

• Data Center Dynamics (2024-05-17)

• CoreWeave: Data Center Regions, Locations, and GPU Cloud

• Dgtl Infra (2024)