Watch on YouTube
Watch on Vimeo
CTERA’s Intelligent Data Platform is designed to unify unstructured enterprise data into a secure fabric that serves as a foundation for AI agents. By transforming the file system into an “agentic coordination layer,” CTERA enables organizations to make their data AI-ready without requiring extensive migration or data movement. CTERA has a long history in the market, initially building a scalable and secure global file system before evolving it into a unified unstructured data fabric that integrates seamlessly with AI agents. The company’s core strengths lie in its deep security expertise, flexibility in deployment across on-premise, hybrid, and cloud environments, and proven global scale, managing hundreds of petabytes of data for complex enterprise and defense customers.
The solution is based on a two-part architecture, featuring the CTERA portal as a software-defined infrastructure that can run in any cloud or on-premises, connecting to object storage buckets from vendors such as Amazon and Azure, as well as private offerings. Data from edge locations is managed by CTERA Edge Filers, lightweight virtual machines that replace traditional NAS solutions, or the CTERA Drive for laptops, ensuring global synchronization. Beyond this core, CTERA has developed new offerings focused on AI readiness, including CTERA Insight for crawling metadata and audit logs, full-text search, CTERA Classify, and CTERA Expert, an embedded AI RAG platform. CTERA emphasizes that it doesn’t provide object storage itself but serves as the “brain” atop existing object storage, integrating with a rich ecosystem of cloud providers, object storage vendors, security solutions, hyper-converged technologies, and various LLMs and AI agents.
CTERA primarily targets cold and warm unstructured data workloads, evolving from NAS replacement to supporting a range of AI applications. Recent innovations include integration with N8N workflows, CTERA Fusion Direct, which allows plugging existing object storage buckets into their data fabric without migration, and CTERA Insight AI, an advanced platform with built-in AI agents that enable users to query data, security audit logs, and make informed decisions. This journey signifies a shift from merely unifying and protecting data to actively understanding and activating it, transforming the file system into an “agentic coordination layer” for AI. A real-world example shared was a maritime customer using CTERA’s fabric to securely and rapidly transfer large volumes of edge-generated data (on ships) over high-latency links to data centers for immediate AI analysis, demonstrating the platform’s critical role in operational efficiency and intelligence gathering.
Personnel: Oded Nagel
Watch on YouTube
Watch on Vimeo
CTERA’s Intelligent Data Platform is designed to unify unstructured enterprise data into a secure fabric that serves as a foundation for AI agents. By transforming the file system into an “agentic coordination layer,” CTERA enables organizations to make their data AI-ready without requiring extensive migration or data movement. The core idea is that files are the natural interface for agents to communicate and collaborate, serving as memory and a workspace. This approach addresses the challenges of unstructured, scattered, and often messy enterprise data, which is typically expensive and inefficient for AI agents to parse directly.
To achieve this, CTERA’s solution automatically generates “derivative artifacts”, such as JSON files, markdown summaries, vector embeddings, and textual representations, which are stored alongside the original files in a `.meta` folder. This process moves much of the reasoning work from inference time to ingestion time, making agent operations much more token-efficient, deterministic, and scalable. When a file is modified, a real-time message bus wakes up, triggering updates to these artifacts. Agents can then efficiently access this structured, summarized data, drastically reducing egress costs and the computational effort required to process large binary files, such as videos or complex documents. The file system acts as a “whiteboard” where agents read and write these artifacts, fostering decoupled communication.
Furthermore, CTERA enhances file system governance for AI agents by enabling fine-grained access control (ACLs) using non-human identities and maintaining audit logs. The platform supports a “bring your own LLM” model, enabling customers to choose different foundation models based on use case, sensitivity, or cost, from cloud LLMs to on-premise solutions. CTERA’s global file system technology with edge caching also facilitates running agents across diverse locations, from on-premise sites to the cloud, providing accelerated local access to data. This file system-centric architecture supports the current trend of code-executing agents by providing a robust, efficient, and deterministic layer for their interactions, leveraging decades of file system development for collaboration and permissions.
Personnel: Aron Brand
Watch on YouTube
Watch on Vimeo
CTERA’s Intelligent Data Platform is designed to unify unstructured enterprise data into a secure fabric that serves as a foundation for AI agents. By transforming the file system into an “agentic coordination layer,” CTERA enables organizations to make their data AI-ready without requiring extensive migration or data movement. Aron Brand, CTO of CTERA, explained that the platform’s content-aware file system goes beyond merely understanding bytes and blocks to understanding the actual content. This enables efficient data preparation for AI agents by removing low-quality data and tagging risky information, thereby streamlining agent operations and eliminating the need for extensive data migration or context rebuilding. CTERA aims to provide an “operating system for agents,” allowing them to reliably understand, trust, and act upon existing enterprise data at petabyte scale.
The CTERA Intelligent Data Platform leverages three core headless services: CTERA Search, CTERA Classify, and CTERA Experts. CTERA Search monitors the global file system for new or modified files, extracting content using various file-type-specific extractors. It then indexes these files into a hybrid full-text and vector database, facilitating robust retrieval based on both data and metadata. This scalable process, built on Kubernetes, continuously indexes data in the background, with discovery tools guiding focus to high-value folders. CTERA Classify builds upon this by labeling and enriching data, using LLMs and other AI models to extract specific schemas from unstructured text (e.g. patient names and dates from medical documents). This derived information is stored as searchable metadata and alternate data streams, which are readily accessible to agents. Finally, CTERA Experts serves as a reasoning layer, enabling agents to answer natural-language questions by querying the underlying search and classified data and acting as a sub-agent within broader agentic platforms. This suite of content services focuses on data curation and preparation for AI, provided as a privately hosted system, not a SaaS offering.
To further enhance data accessibility and integration, CTERA introduced Fusion Direct, a solution for making existing object storage data AI-ready without migration or format changes. Fusion Direct connects directly to S3 buckets, providing file access while maintaining the original objects as the gold copy. This creates a unified object and file view, beneficial for HPC and AI training workloads, allowing CTERA’s content services to enrich data directly within these buckets while leveraging caching capabilities. The platform also announced a new connector for N8N, a low-code/no-code automation platform. This integration allows users to build visual workflows and pipelines that interact with CTERA data, accessing its content services, including enriched metadata, semantic search, and experts, directly from N8N’s rich ecosystem of connectors. These developments underscore CTERA’s strategy to provide versatile access and automation methods for AI-ready data.
Personnel: Aron Brand
Watch on YouTube
Watch on Vimeo
CTERA’s Intelligent Data Platform is designed to unify unstructured enterprise data into a secure fabric that serves as a foundation for AI agents. By transforming the file system into an “agentic coordination layer,” CTERA enables organizations to make their data AI-ready without requiring extensive migration or data movement. This capability addresses the challenges faced by enterprises across financial services, telecom, and manufacturing, which often grapple with petabytes of disorganized unstructured data that needs to be leveraged for business intelligence and automated workflows. The platform facilitates this by allowing agents to automatically process files dropped into designated “magic folders” and extract valuable insights.
The platform’s Content Services are key to this process, transcribing audio files, performing OCR on scanned PDFs, classifying data, and extracting or augmenting metadata. Users define schemas, often in JSON format, to specify the exact information to be extracted from unstructured data, such as call outcomes, customer complaints, or contract details. CTERA then automatically generates “derivative artifacts”: summaries, tags, keywords, and structured metadata files that inherit the permissions of the original file and are stored alongside it. This approach allows AI agents and large language models (LLMs) to efficiently query and analyze data as if it were a structured database, without needing to directly access the original, often complex, raw binary files.
CTERA demonstrated several real-world applications, including analyzing call center conversations to identify churn risk and escalate relevant calls, processing legal contracts to extract critical information, and classifying sensitive data, such as Protected Health Information (PHI), to enforce access restrictions. The system supports building autonomous agents that can execute these workflows in real time or periodically, and even generate dynamic dashboards. Addressing security and governance concerns, CTERA ensures that all derivative artifacts maintain the security posture of the original data, integrates with data loss prevention (DLP) tools, and offers immutable snapshots as frequently as every five minutes to protect against accidental deletions or malicious actions by agents or users, thereby providing robust audit trails and recovery capabilities.
Personnel: Dylan Locsin
Watch on YouTube
Watch on Vimeo
CTERA’s Intelligent Data Platform is designed to unify unstructured enterprise data into a secure fabric that serves as a foundation for AI agents. By transforming the file system into an “agentic coordination layer,” CTERA enables organizations to make their data AI-ready without requiring extensive migration or data movement. The new CTERA Insight AI system extends this by providing operational intelligence for IT administration, moving beyond supporting customer use cases to streamlining internal IT tasks. It addresses common pain points such as a lack of understanding of existing data, unknown risks posed by legacy permissions, and difficulties prioritizing actions in large, growing unstructured data estates.
CTERA Insight AI functions by ingesting two primary data streams: file metadata and all activity logs (reads, writes, modifications, deletes) from an organization’s global file system and storage platform. This information is stored in a searchable data lake, accessible for analysis. A key innovation is the use of AI capabilities, including LLM knowledge, semantic parsing, and OCR, to provide a deep, semantic understanding of data content, going beyond mere file names or traditional metadata. This allows the system to differentiate between, for example, a call center recording and a song, even if both are MP3s, enabling more accurate insights into data type, purpose, and potential risks like sensitive data exposure or copyright violations.
The platform provides practical operational benefits, such as identifying candidates for archiving or deletion to optimize storage costs (e.g., locating 116 terabytes of data more than five years old with no recent access). For security and governance, it monitors both human and AI agent activities, treating agents as human users with non-human identities, enabling auditing of unusual access patterns, mass deletions, or potential data leaks. Users can leverage natural language queries to generate custom reports and views, eliminating the need for predefined dashboards and significantly accelerating investigation times from hours to seconds. This empowers IT professionals with actionable insights for data lifecycle management, compliance, and cyber protection, making the file system truly intelligent and manageable.
Personnel: Aron Brand, Dylan Locsin
Thank you for being part of the Tech Field Day community! Our mailing list is a great way to stay up to date on our events and technical content, and we appreciate your signup.
We promise that we’ll never spam you, send ads, or sell your information. This list will only be used to communicate with our community about our events and content. And we’ll limit it to no more than one message per week.
Although we only need your email address, it would be nice if you provided a little more information to help us get to know you better!
