With Forward Deployed Engineering (FDE), data management specialist CTERA embeds its own experts in enterprise teams to take AI use cases on unstructured file data into production, as shown on IT Press Tour.
CTERA is adding a hands-on service to its software business: with Forward Deployed Engineering (FDE), the company places its own engineers inside customer organizations to take specific AI use cases into production on the customer’s unstructured data, according to the vendor.
The announcement, made on 6 October 2026 and presented to journalists a day earlier at the IT Press Tour, targets a problem that vendors, analysts and IT departments describe in increasingly similar terms: AI tools are widely available, but reliable results in production are not. CTERA cites a Futurum survey of 818 data leaders from the first half of 2026, according to which a shortage of specialized talent is the fastest-rising factor behind failed AI projects. The share naming insufficient skills and expertise as an organizational challenge more than doubled within six months, from 4.8 to 10.4 percent.
CTERA argues that the problem is sharpest with unstructured file data. Estates built up over decades are often duplicated, stale and inconsistently permissioned. Pointing AI at such data without preparation can produce confident but wrong answers, or surface content to people who should not see it. “Enterprises aren’t short of AI tools,” said CEO Oded Nagel. “What they’re short of is the time and specialized skills to connect those tools to their own data and workflows safely.”
Nagel framed the issue as a data problem rather than a model problem. Adoption is near-universal, he said, with nearly 90 percent of enterprises using AI, yet most do not get beyond pilots; large banks, government agencies and healthcare providers in particular are not yet in production. Security adds to the pressure: the company’s slides cite a survey in which 67 percent of executives believe their company has already suffered a breach through unapproved AI tools, and a report that GenAI data movement rose 80 percent year on year, bypassing legacy data loss prevention. CTERA’s answer is to move AI to the data rather than data to AI, which, it says, avoids ungoverned copies and keeps file permissions intact.
According to the company, each engagement is staffed by a small team of a Forward Deployed Engineer and a Deployment Strategist, who work as part of the customer’s project team with access to its environments and subject-matter experts. The approach runs in four phases: discovery and data grounding, a first production use case, scaling to further workflows and sites, and handover for independent operation. Discovery reads the file estate by metadata only. In the session, executives said engagements typically start with at least a week of preparation and often run for around a month; larger customers might assign a permanent engineer.
CTERA names four principles. Work starts from the data: engineers assess volumes, access patterns and source systems to validate workflows with evidence. Data is prepared for safe use: relevant data sets are classified, organized and governed in place, and AI only works with content each user is already authorized to access. Solutions are built into existing workflows through open standards such as the Model Context Protocol (MCP) and automation platforms such as n8n, and are validated against real data before go-live. Finally, ownership is transferred: customer engineers join from the first phase, configuration is documented as it is built, and an engagement is complete once the customer can add the next use case without CTERA.
FDE builds on the CTERA Intelligent Data Platform, which, according to the company, provides AI-governed access to file data across edge locations, data centers and cloud environments without moving it. CTO Aron Brand pointed to a medical-legal customer as an example: a small firm turned scanned, handwritten medical files into case narratives for legal review using a CTERA Classify pipeline built with engineering support. Other reference projects shown in the session included a manufacturer analyzing more than one petabyte of metadata across four regions and a financial services company making about 27 terabytes of mortgage records searchable. Brand cautioned that mixed, low-quality data leads to convincing but wrong answers.
The step marks a shift for a vendor that, as Brand explained, used to treat professional services as something to minimize. Customers now ask for hands-on help, he said. Around ten people currently work in the FDE team, which can be scaled. Nagel stressed that CTERA does not intend to become a services business; its strength remains the technology. Brand added that the engineers will also touch non-CTERA systems such as Salesforce where a use case requires it.
FDE is available now for organizations using the CTERA Intelligent Data Platform. Engagements can be structured by time or as fixed-scope projects with milestone-based acceptance criteria. Independent customer results for the new service are not yet available; the examples and figures come from CTERA and the surveys it cites.
Dr. Jakob Jung is Editor-in-Chief of Security Storage and Channel Germany. He has been working in IT journalism for more than 20 years. His career includes Computer Reseller News, Heise Resale, Informationweek, Techtarget (storage and data center) and ChannelBiz. He also freelances for numerous IT publications, including Computerwoche, Channelpartner, IT-Business, Storage-Insider and ZDnet. His main topics are channel, storage, security, data center, ERP and CRM.
Contact via Mail: jakob.jung@security-storage-und-channel-germany.de
My experience with this method differed when dealing with legacy databases.