Building an AI system is one challenge. Getting it to work inside a business is another.
Enterprise AI deployments often involve more than integrating a model or building a new application. They require engineers to work with existing software, understand business processes, and adapt technology to each customer’s operational realities. As companies move AI from pilots into production, this is creating demand for a different kind of engineering role.
Forward-deployed engineers (FDEs) sit at this intersection of software development, product engineering and customer operations. They work directly with businesses to make technology work within their existing environments, often building solutions that a standardised product alone can’t deliver.
The growing importance of this role is reflected in Enmovil’s hiring plans. The agentic AI company, which builds software for supply chain and logistics orchestration, plans to expand its FDE team from around 30 employees to 100–120 within a year. The move comes as it scales enterprise AI deployments and expands into the US and Southeast Asia.
The company currently has 175 employees, including around 100 in development. It expects revenue to increase from approximately $5 million to $8–9 million by the end of the current financial year. Enmovil has raised around $8 million to date, including a $6 million Series A round approximately a year ago.
“As systems and technology become more complex, teams need to work closely with customers, understand their operations and make the technology practical for them to use. Enterprise solutions do not follow a one-size-fits-all model,” Ravi Prasad Bulusu, co-founder and CEO of Enmovil, tells AIM.
For Enmovil, the expansion reflects the need to deploy AI across interconnected supply chain processes, where technology must account for customer-specific workflows and operational constraints.
Why FDEs Are Becoming Central to Enterprise AI
Unlike traditional software development teams that primarily build products centrally, FDEs work at the intersection of engineering, product development and customer operations. They help adapt technology to specific operational requirements, often working alongside customers to integrate software into existing systems.
For enterprise AI deployments, this becomes particularly important when systems need to connect with enterprise resource planning platforms, warehouse management systems, transport management systems and other applications. The technology must also account for customer-specific workflows, operational constraints and approval requirements.
Enmovil’s platform connects supply chain planning, inventory management, procurement, replenishment, and logistics execution through an agentic AI layer. Its customers span manufacturing, automotive, FMCG, e-commerce and direct-to-consumer businesses, including Maruti, Mahindra, TVS Motor, Volvo, Daimler, Nestlé, ITC and The Whole Truth Foods.
Depending on the use case, the company competes with supply chain technology providers such as SAP, Blue Yonder, Kinaxis, o9 Solutions and Pando. Enmovil differentiates itself by bringing planning and execution together through an AI-driven platform rather than focusing on individual supply chain functions in isolation.
That approach also creates an implementation challenge. Enterprises have different systems, data structures and operational processes. FDEs are expected to bridge the gap between the platform’s capabilities and the specific needs of each customer.
Before starting Enmovil, Bulusu spent about 20 years working in technology, including as a director at NVIDIA and Intel. He also worked at PortalPlayer, a startup later acquired by NVIDIA. His co-founders, Nanda Kishore, COO, and Venkat Moganty, CTO, also bring technology experience to the business.
An AI Workforce Needs a Human Engineering Workforce
Enmovil describes its platform as an AI workforce that coordinates supply chain activities through individual agents. At the centre is CADDIE, a supervisory agent that communicates with other agents and orchestrates workflows across planning and execution.
A supply chain manager could ask CADDIE to assess demand and fulfil orders. The system would check available inventory, prioritise customers, identify replenishment requirements, and determine how to deliver goods. It can also bring humans into the loop when approvals are required.
“We have uniquely built an agentic platform that can communicate with individual modules, each of which can function as an agent, and coordinate them into workflows that execute supply chain activities. This effectively provides customers with an AI workforce,” Bulusu says.
However, deploying such a system requires more than making agents communicate with one another. The platform must integrate with existing enterprise systems and reflect how a customer’s supply chain actually operates.
Bulusu says the company’s optimisation engine remains its core technological differentiator, while the agentic layer makes the system easier to use. FDEs help ensure the technology is configured and adopted in ways that address specific business problems.
This places customer-facing engineers in a key role as enterprise AI moves from experimentation towards operational deployment. Their work involves understanding the customer’s environment, aligning the platform with existing processes and helping businesses integrate AI-driven decisions into day-to-day operations.
From Supply Chain Planning to Execution
Enmovil’s platform supports demand forecasting, inventory and replenishment planning, logistics network optimisation and dispatch. It operates as an intelligence layer over existing systems such as SAP, Oracle and Blue Yonder, rather than replacing them.
The company claims its platform has improved forecast accuracy by 30–40%, reduced dispatch costs by 10–15% and cut manual effort by more than 75%. Bulusu also says customers can begin seeing value in less than six weeks from purchase order to delivery.
“Our USP is that we provide a complete agentic operating system for the supply chain. From assessing inventory to coordinating dispatch, the platform brings the process together in one place, with the appropriate approval loops,” he says.
For e-commerce and D2C businesses, connecting forecasting, inventory, replenishment, and logistics matters because demand fluctuations and delivery constraints directly affect customer experience. Enmovil’s approach aims to coordinate these decisions rather than leave them distributed across separate systems.
The company has more than 40 customers in India and is beginning to expand internationally. As its customer base grows, the planned FDE expansion is expected to support the work required to adapt and deploy its platform across different enterprise environments.
The Enterprise AI Hiring Shift
Enmovil’s hiring plans reflect a broader challenge facing enterprise AI companies: building the technology is only one part of adoption. Businesses also need engineers who can translate AI capabilities into practical workflows, integrate systems and work closely with customers.
The company expects its FDE team to grow significantly faster than the rest of its workforce, increasing from around 30 employees to 100–120 within a year. That would make the function a substantially larger part of the organisation as Enmovil expands its enterprise deployments.
The shift does not mean software development becomes less important. Enmovil still has around 100 employees in development. Instead, its hiring priorities suggest that product engineering and customer-facing implementation will need to scale together.
For Enmovil, the next phase of growth will depend not only on its agentic platform’s capabilities, but also on how effectively its engineers can make it work within complex enterprise operations.