Large IT services firms are acquiring specialist engineering companies to strengthen their positions in markets being reshaped by artificial intelligence (AI). But owning domain expertise and knowing how to apply AI to complex engineering work are two different things. KPIT Technologies, an automotive engineering services company, offers a useful example of this challenge.
The Pune-headquartered company is embedding AI across its engineering and delivery operations, with management citing improvements in productivity, delivery speed, and rework. Yet its revenue grew just 0.2% year-on-year (YoY) in constant currency in Q2 FY27, even as its EBITDA margin rose to 17.8%.
KPIT’s experience raises a question for larger IT firms pursuing similar capabilities through acquisitions: can they turn specialist expertise into new AI-led business, or will the immediate gains primarily come from more efficient delivery?
The Changing Economics of Engineering Services
The distinction between using AI to improve productivity and using it to generate new business came up during KPIT’s Q2 earnings call. Responding to a question from AIM on whether AI was helping KPIT win new engagements or primarily improving productivity in existing projects, Co-founder and CEO Kishor Patil pointed to its role in margin improvement.
COO Anup Sable described a shift towards what KPIT calls AI-first projects. He said they were resulting in faster delivery and higher productivity, with “almost zero rework”. KPIT has also finalised the infrastructure, domain context, and supporting systems required to scale these interventions across projects.
The immediate benefit is operational efficiency. However, these gains do not necessarily translate to higher revenue. If a project requires fewer engineering hours, the services provider may also have to rethink how it prices and delivers the work.
KPIT has not disclosed how much AI contributed to its Q2 margin improvement, while stating that “AI-first projects” contributed roughly 9% of total revenue. The company attributed the margin expansion primarily to operational efficiency and a better revenue mix.
Domain Expertise Remains Paramount
The next stage of AI adoption could make specialist knowledge more important in engineering services. Automotive engineering involves complex systems, including embedded software, vehicle architectures, and safety-critical applications. Applying AI in these environments requires an understanding of how different components work together.
On how AI was changing competition between specialist engineering companies and larger IT services firms acquiring similar capabilities, Patil argued that industry-focused companies have an advantage in delivering “deterministic outputs” for enterprise clients.
Domain depth and experience in putting systems together remain important, he said. He also pointed to the growing traction of AI companies focused on developing harnesses, and the role of partnerships with frontier labs such as Anthropic in helping companies serve enterprise clients.
This requirement extends beyond automotive engineering. Enterprise modernisation also requires an understanding of existing systems and their dependencies. Mphasis CEO Nitin Rakesh has argued that AI needs to go beyond superficial changes to legacy IT.
Acquiring Specialist Capabilities
This helps explain why larger IT services companies are acquiring businesses in engineering, automotive, semiconductors, and AI. These deals provide access to specialist talent, technology, and established customer relationships. They also give larger firms an entry into markets that require capabilities they may take years to develop internally.
Persistent Systems’ acquisition of Nagarro is one example of the push to expand digital engineering capabilities and scale. HCLSoftware has also used acquisitions to strengthen its data and AI portfolio, including its purchases of Wobby and Jaspersoft.
But acquiring a specialist company does not automatically transfer its expertise across a larger organisation. Buyers must retain specialised talent and integrate the acquired teams into their existing delivery operations.
Gaurav Vasu, Founder and CEO of business intelligence firm UnearthInsight, sees a risk in assuming that acquisitions will translate into stronger growth. “When larger IT firms acquire smaller specialised companies, it is mostly about adding revenue in the short run, and maybe some capability. That does not guarantee growth,” he says.
Vasu also points to the risks of buying businesses whose customers operate in industries facing disruption or margin pressure, including automotive and retail. Stress in these sectors affects their technology vendors as well, making it harder for acquisitions to deliver the expected returns.
The Test of Delivery Efficiency
KPIT’s new engagements provide some indication of the demand for its engineering capabilities. The company secured $213 million in total contract value during Q2, covering electric powertrains, vehicle engineering, connected vehicles, diagnostics, autonomous systems, and cybersecurity.
However, revenue growth remains constrained by spending cuts and declines in its major customer accounts. Management expects improvements in revenue mix, growth and delivery efficiency to support profitability, and is targeting an EBIT margin of 16% in early FY28.
The rise of boutique AI firms in India adds to the competitive pressure. These companies are building specialised AI capabilities and competing for the same engineering and enterprise transformation work that larger IT services firms want to capture.
For larger firms, acquisitions will have to deliver more than additional capabilities or improvements in delivery efficiency. They must help win new AI-led programmes, expand existing engagements, and open up sources of revenue that would otherwise be difficult to access.
