EXL to Acquire iMerit for Up to $310M, Expanding Enterprise AI Capabilities

Artificial intelligence has entered a new phase. The conversation is no longer about building the biggest model. It is about building the model enterprises are willing to trust.

ExlService Holdings, Inc. (NASDAQ: EXLS) has announced a definitive agreement to acquire San Jose-based iMerit in a transaction valued at up to $310M. The deal includes $170M in upfront consideration and up to $140M in milestone-based payments over the following 2 years. The acquisition is expected to close during Q3 2026, subject to customary closing conditions and regulatory approvals.

The acquisition pairs EXL's enterprise data and AI platform with iMerit's expertise in AI model training, reinforcement learning from human feedback (RLHF), model evaluation, and high-quality human-in-the-loop workflows. Together, the companies are positioning themselves to address one of enterprise AI's biggest challenges: moving from experimentation to reliable production systems.

This is more than another AI acquisition. It is a signal that the infrastructure surrounding AI models has become just as valuable as the models themselves.

What Happened

ExlService Holdings, Inc., headquartered in New York City, signed a definitive agreement to acquire iMerit through its subsidiary, Clairvoyant AI, Inc. The acquisition is valued at up to $310M, consisting of $170M at closing and as much as $140M in milestone-based consideration over 2 years. The transaction is expected to close during Q3 2026 following customary regulatory approvals and closing conditions.

The transaction strengthens EXL's Data & AI strategy by adding capabilities in foundation model training, AI model evaluation, RLHF, and expert human feedback. EXL Chairman and CEO Rohit Kapoor described the acquisition as another step toward helping enterprise customers move from AI experimentation into production-grade deployment.

Founder and CEO Radha Ramaswami Basu built iMerit into a specialized AI data company focused on training, evaluating, and improving AI systems across multiple data modalities, including text, images, video, audio, medical imaging, and LiDAR. That specialization has become increasingly valuable as enterprises demand trustworthy AI rather than impressive demonstrations.

Why This Acquisition Matters

Enterprise AI has quietly developed an uncomfortable truth: large language models receive the headlines, but data quality, model evaluation, and human expertise determine whether those models survive contact with the real world. That invisible layer is where iMerit built its business.

Through its Ango Hub platform, iMerit provides AI data workflows, model evaluation, quality assurance, workflow automation, and reinforcement learning capabilities. The company's Scholars network adds another competitive advantage by connecting domain experts, including physicians, engineers, scientists, linguists, and legal professionals, to complex AI evaluation tasks where accuracy matters more than speed.

Enterprise customers increasingly understand that a model is only as reliable as the data, evaluation, and expert feedback behind it. That realization is reshaping acquisition strategies across enterprise AI.

The Strategic Fit Between EXL and iMerit

EXL already operates at significant scale across healthcare, insurance, banking, financial services, and other regulated industries where data quality and compliance are business requirements rather than marketing slogans. Its AI portfolio includes EXLerate.ai, EXLdata.ai, and EXLdecision.ai.

Adding iMerit's capabilities extends that platform with AI model training, RLHF, evaluation, and expert data operations that previously required specialized external providers. The acquisition also expands EXL into adjacent AI markets, including autonomous mobility, robotics, physical AI, and multimodal AI systems, where expert-labeled data remains a critical competitive advantage.

Building those capabilities organically would likely require years of recruiting specialized talent, developing production workflows, earning customer trust, and refining quality systems. Acquiring them compresses that timeline dramatically.

Market Context: AI Infrastructure Is Becoming the Real Battleground

The AI market has matured enough that enterprises are asking different questions: Can this system be audited? Can regulators trust its outputs? Can it handle domain-specific knowledge? Can it operate safely in production?

Those questions cannot be answered with larger models alone. They require evaluation frameworks, reinforcement learning, expert reviewers, governance processes, and high-quality training data. That is precisely where enterprise spending is beginning to concentrate.

Rather than assembling six specialized vendors to manage data preparation, annotation, evaluation, deployment, monitoring, and optimization, enterprises increasingly want integrated platforms capable of supporting the entire AI lifecycle. EXL's acquisition of iMerit reflects that broader shift toward platform consolidation.

A Different Kind of Startup Story

Perhaps the most interesting part of this transaction is not the purchase price. It is how iMerit reached it.

The company reportedly raised approximately $24M in outside capital before agreeing to a transaction valued at up to $310M. Instead of chasing perpetual fundraising, iMerit focused on building a profitable business around an increasingly difficult technical problem while developing proprietary expertise that larger companies could not easily reproduce.

In today's startup ecosystem, fundraising often receives more attention than execution. Markets eventually correct that imbalance because customers pay for outcomes, acquirers pay premiums for scarce capabilities, and investors reward companies that compound value instead of simply extending runway.

What This Signals for Enterprise AI

This acquisition illustrates a larger shift taking place across enterprise AI. Competitive advantage is moving away from simply owning powerful foundation models and accumulating around the systems that make those models reliable, governable, explainable, and deployable inside complex organizations.

Trust has become infrastructure. Organizations that can combine enterprise relationships, proprietary industry knowledge, AI platforms, model evaluation, and expert human feedback will be positioned to capture a growing share of enterprise AI budgets.

EXL's acquisition of iMerit represents a strategic investment in that future. The next phase of AI will not be won by the loudest announcements. It will be won by the companies that quietly make artificial intelligence dependable enough for businesses to build on.