Verda Secures $189M Series B, Plans S3-Compatible Object Storage for AI Infrastructure

AI cloud startup Verda recently announced the successful closure of its Series B funding round, securing an impressive $189 million. This latest injection of capital brings Verda's total funding to over $450 million and is earmarked for significant expansion, including increasing data center capacity and deepening investment in its platform. A key strategic initiative highlighted in the announcement is the company's plan to release S3-compatible object storage. This development is highly significant for practitioners in cloud, DevOps, and AI. The emergence of specialized AI cloud providers like Verda, focusing on high-performance infrastructure tailored for AI workloads, indicates a maturation of the AI ecosystem. The commitment to S3-compatible object storage is particularly important. S3 has become a de facto standard for object storage, offering a familiar and robust API for developers and applications. By adopting S3 compatibility, Verda is not only providing a scalable storage solution but also ensuring easier integration with existing tools and workflows that already interact with S3. This reduces the friction for AI developers and data scientists who can leverage their current knowledge and tooling to manage large datasets within Verda's AI-optimized environment. This move by Verda aligns with a broader trend in the cloud and AI landscape where object storage is increasingly recognized as a critical component for AI workloads. Traditional storage solutions often struggle with the scale, performance, and cost requirements of AI training and inference. Cloud providers have been enhancing their object storage offerings with features like high-performance tiers and AI-specific optimizations. For instance, Google Cloud introduced Cloud Storage Rapid to address bottlenecks in AI jobs, offering extreme throughput and ultra-low latency while maintaining object storage's durability and scalability. Similarly, AWS has been evolving S3 with features like S3 Files to make object storage more accessible to file-based applications, including those used in AI and ML. The need for efficient data handling for AI is also driving innovation in Kubernetes storage, with solutions like NetApp Trident 26.10 supporting Container Object Storage Interface (COSI) for managing S3-compatible object buckets within Kubernetes. This collective movement underscores the imperative for object storage to evolve to meet the demanding needs of AI, from data ingestion and preparation to model serving and long-term archival. In practice, this means that practitioners should closely watch how these specialized AI cloud providers differentiate their S3-compatible offerings. While S3 compatibility ensures a baseline of interoperability, the underlying performance characteristics, pricing models, and integration with other AI-specific services (such as GPU instances or AI/ML platforms) will be crucial decision factors. Developers should assess whether the performance gains and specialized features offered by platforms like Verda outweigh the potential benefits of using general-purpose cloud object storage from hyperscalers. Furthermore, the emphasis on S3 compatibility suggests that investing in skills and tools around the S3 API and related data management practices will continue to be valuable for anyone working with large-scale AI data. As Verda aims to deploy Nvidia Vera Rubin NVL72 systems and offer access to Arm's AGI CPU, the integration of their S3-compatible object storage with such cutting-edge compute resources will be a key area for practitioners to evaluate for optimizing their AI pipelines.