As businesses move beyond isolated AI experiments toward enterprise‑wide adoption, many are rethinking their team structures and strategies to foster sustainable growth.
In this episode of the HCLTech Trends and Insights podcast, we sat down with Srinivas Kompella, Senior Vice President of Data and AI Enterprise at HCLTech, to unpack the findings of HCLTech’s latest research on building an AI‑led operating model. Kompella brings deep expertise and firsthand insights into what’s working, and what’s not, as organizations navigate this critical shift.
The multidisciplinary imperative
Enterprises that have seen early success with AI often began with small, contained projects. But Kompella notes that scaling AI requires a much broader set of capabilities. “As enterprises begin to shift from experimentation, small use cases to broader adoption of AI, the realization that is coming in is that it is a truly multidisciplinary team that will enable scaling of AI.”
He explains that such a team must encompass:
- Robust infrastructure foundation, which underpins both performance and cost‑effectiveness
- World‑class data engineering and data science skills to establish a rock‑solid data foundation
- AI expertise, not only in using large language models, but in designing full‑scale AI systems, especially as Agentic AI gains prominence
- Business process knowledge, since “the purpose of scaling AI is to positively impact business,” whether that’s growing top line or transforming customer experience
By aligning these disciplines around specific business functions, such as sales and marketing, organizations can move from one‑off pilots to repeatable, standardized solutions.
“Productization…builds that level of repeatability, standardization and consistency,” said Kompella, “making a product‑operating model an absolute must” for any enterprise seeking truly transformational impact.
Balancing adaptation with governance
With generative and Agentic AI evolving at breakneck speed, CIOs and CXOs often worry about how to stay agile without compromising on safety or accountability. Kompella points to the multidisciplinary product team as the answer. Our research highlights that firms that adopt a product-aligned approach to achieve it are 4x more likely to realize strong ROI outcomes.
“The moment we bring in a multidisciplinary team…they’re able to evaluate and assess the impact” of emerging technologies, deciding what’s relevant and safe. He underscores the importance of a solid technical foundation: “If this foundation is architected…the changes in AI technologies can actually be better evaluated and put to purpose [at pace].”
Yet evaluation alone isn’t enough. Responsible AI must be baked into the governance framework, from centralized oversight down to citizen developers at the edge. “Most users...will be building AI at the edge, on their laptops and on their devices,” said Kompella. “That’s going to mean…incredible levels of governance, not just from an AI technology standpoint, but also in terms of access to data and insights.”
At a higher level, an AI Center of Excellence, or council, can provide the steering needed to vet new tools for safety, applicability and ROI. “This governance mechanism,” added Kompella, “will ensure that the technologies that are coming in are deployed in the right way for better ROI.”
Building trust through consistency
Trust is the currency of AI adoption. “One of the biggest factors in scaling AI is building trust,” emphasized Kompella. He draws a parallel to everyday consumer experiences: “If you’re using any GPS maps application, you’re trusting that AI to say…take this route versus that route.” Today’s users expect that same reliability from enterprise AI systems, and they’re quick to abandon tools that fail them.
To earn that trust, AI must not only match human accuracy but surpass it. “To err is human. Unfortunately, that doesn’t apply to AI,” he noted. When AI products consistently deliver correct results, organizations can gradually extend responsibility to “citizen AI developers” while maintaining confidence in their systems.
Overcoming team‑structure bottlenecks
Legacy organizational models, including functional hierarchies or matrix structures, often create silos and slow decision‑making. Digital‑native companies, by contrast, form agile teams around specific products, launching them quickly and iterating in real time.
In a product‑aligned operating model, a dedicated product owner drives cross‑functional team accountability, even as members remain in their respective reporting lines. “Accountability is driven by the product team,” explained Kompella, “as opposed to the functional alignment.” While this shift demands robust change management and leadership alignment, it also opens new career pathways and fosters broader skill development. While only a minority of organizations believe they’ve achieved full alignment, 95% agree that a shift in leadership mindset is critical — suggesting that the product-aligned model is not just a structural shift, but a strategic leadership priority.
Agility through problem‑centric data strategies
Traditional data architectures focused on centralized repositories, such as enterprise data warehouses designed for reporting. But AI’s promise lies in solving discrete business problems with agility.
“Within the age of AI…you can’t take months or years to build a central data repository,” said Kompella. Instead, teams should work backward from a clear use case, such as improving campaign effectiveness, to only identify the data and infrastructure needed to deliver results. This problem‑centric approach reduces time to value and reinforces the product‑operating mindset.
Cultivating a culture of experimentation
Finally, Kompella underscored the role of culture in driving continuous innovation. “The technology is there…it’s a function of imagination on how to apply the technology that needs to be applied.” CXOs must strike the right balance between governance and creative freedom.
He points out that less than 5% of AI experiments make it into production today, which is a stark reminder of the gap between ideation and impact. By channeling experimentation through well‑governed product teams, enterprises can convert more ideas into measurable returns.
“The job of governance is less about control and more about how to channel that energy into the right direction,” said Kompella, “because the technical talent and technologies are there. The moment that channeling starts happening, the innovation will start converting into returns on investment.”
With a product‑aligned operating model, organizations can unite infrastructure, data, AI and business process expertise under clear accountability, embedding governance without stifling creativity. As Kompella’s insights reveal, this multidisciplinary framework offers the repeatability, trust and agility needed to scale AI from isolated pilots to enterprise transformation.
