When Tata Consultancy Services (TCS) said it would build a dedicated cadre of upto 8,900 forward-deployed engineers (FDEs), it appeared to be another hiring move driven by the AI boom. In reality, it signals something much bigger. India’s largest IT services company is positioning itself to compete directly with AI model makers such as OpenAI and Anthropic for the fastest-growing part of the AI value chain: enterprise deployment.
For years, the roles were clearly defined. Foundation model companies built increasingly capable AI models, while IT services firms integrated those models into enterprise systems and managed implementation. That distinction is beginning to blur. Model companies are moving closer to customers, while TCS is pushing deeper into deployment – a role increasingly being claimed by AI vendors themselves. The battle is no longer just about building the best model. It is increasingly about who helps enterprises put AI into production.
Engineers at the frontline
The forward-deployed engineer (FDE) model, first popularised by Palantir, has become central to the go-to-market strategy of companies including OpenAI and Anthropic. Rather than selling software and handing implementation to partners, they embed engineers with customers to redesign workflows, integrate AI into business processes and ensure systems deliver measurable results.
TCS now wants similar capabilities. Chief executive K Krithivasan said the company plans to create an FDE organisation equivalent to around 1 to 1.5% of its workforce. TCS has already built an annualised AI services business worth $2.6 billion, but the next stage of growth will depend less on winning AI mandates and more on helping customers move from pilot projects to enterprise-wide deployments. That is where forward-deployed engineers become valuable.
“The move reflects where enterprise AI is heading,” said Jaspreet Bindra, founder of AI&Beyond. “Having access to a model is no longer enough. The value comes from deploying it inside an enterprise and making it deliver business outcomes. That requires engineers who understand technology, business processes and customer problems.”
Bindra said OpenAI has already built deployment teams while Anthropic is expanding its own forward-deployed engineering capabilities. “Model companies are no longer content to remain upstream providers. They increasingly want to own the deployment layer that has traditionally belonged to IT services firms.”
That changes the equation. If AI companies build the models, deploy them and optimise enterprise workflows themselves, the role of systems integrators inevitably shrinks. TCS’ answer is to build that capability before the model companies can own it.
Unlike conventional software engineers working from offshore delivery centres, forward-deployed engineers spend much of their time with customers. They combine software engineering with consulting and industry expertise, helping organisations redesign processes, integrate AI into existing systems and solve practical business problems.
That marks a departure from the model that made Indian IT globally successful. For decades, the industry’s advantage came from scale, offshore delivery and cost efficiency. AI deployments demand something different: smaller, specialised teams working closely with customers to translate technology into business outcomes.
Harish Mehta, founder and executive chairman of Onward Technologies, sees this as the next phase of the industry’s evolution rather than a disruption. “To justify their valuations, AI model companies will naturally move into enterprise deployment and services,” he said. “Indian IT firms therefore have to move beyond implementation and focus on business transformation. Trust, workflow redesign and domain expertise will matter much more.”
CP Gurnani, co-founder and vice chairman of AIONOS, believes the opportunity lies in owning the “last mile” of enterprise AI. “The real battle was never about who builds the most powerful model. It’s about who can turn that model into something a business can actually measure,” he said. “Model ownership alone won’t decide the future. Whoever can operationalise AI responsibly, at scale, inside real businesses will.”
The new AI race
The competitive advantage is shifting from delivering software at scale to delivering business outcomes. That is why TCS’ announcement matters beyond one company. Indian IT firms have traditionally competed with global consulting firms, enterprise software vendors and cloud providers. Increasingly, they are also competing with the creators of the AI models they deploy.
Every successful deployment strengthens the relationship between AI vendors and enterprise customers while opening up opportunities for consulting, implementation and support revenue. If model companies continue expanding into those services, they will move into an area that has long been the preserve of firms such as TCS, Infosys, HCLTech, Wipro and Tech Mahindra.
Forward-deployed engineering is more than another delivery model. It changes where IT services firms engage with customers – at the point where AI strategy, workflow redesign and implementation come together. TCS is the first Indian IT company to publicly commit to building such a capability at scale, and its peers are likely to watch the experiment closely.
By investing in thousands of forward-deployed engineers, TCS is betting that the biggest opportunity in enterprise AI lies not in building foundation models but in making them work inside large organisations. If that bet succeeds, Indian IT firms may soon find themselves competing as much with OpenAI and Anthropic as with one another.
