
Nitin Garg is a technology and product management professional with more than 20 years of experience developing enterprise systems, leading digital transformation initiatives, and building products that help organizations make better use of their data. His work spans financial technology, cloud computing, artificial intelligence, healthcare, and business operations, with a focus on solving complex technical challenges and turning business requirements into practical technology solutions.
Throughout his career, Garg has held senior product management, technology, and strategy roles, working on initiatives that range from enterprise data integration and business intelligence platforms to AI-powered automation and large-scale financial management systems. His experience includes developing a financial management product responsible for allocating and managing $1.5 billion in annual budgets, delivering enterprise reporting tools used by thousands of employees, and helping build AI-driven workflow automation capabilities that connect applications and services across complex technology environments.
With an engineering background and an MBA from Indiana University’s Kelley School of Business, Garg brings both technical and commercial perspectives to product development. He has worked closely with engineering teams, business leaders, and other stakeholders to address challenges involving data quality, system integration, security, and infrastructure performance. A recurring theme in his work is the importance of understanding how data moves between systems and making thoughtful decisions about which technical investments will deliver the greatest value.
As organizations increasingly incorporate artificial intelligence into their operations, Garg believes the role of product managers is changing. In particular, he sees AI agents and automated workflows creating new demands on enterprise data architecture, requiring product leaders to develop a deeper understanding of the underlying systems that support these technologies.
In this interview with IntelligentHQ, Garg discusses lessons from his career in enterprise technology, shares examples of how he has approached complex data integration challenges, and explains why strong data foundations will become even more important as businesses move toward AI-driven operations.
Throughout your career, you’ve worked across product management, engineering, and enterprise technology. How has that experience shaped the way you approach building products that depend on data from multiple systems?
I prefer to operate on first principles, a mental model that breaks down complex problems into their most basic forms and builds solutions from scratch.One such architecture design principle I use for designing products with data integration across multiple systems is to separate the “core system” system architecture of the product and couple it with an “interface/integration layer”. This layer manages the services for all external data exchanges and the core system can call upon this integration layer for relevant data. This integration layer ensures that all corresponding source data is refreshed independently in the right order with the right dependencies from the source system with appropriate master data management governance.
Large organizations often operate with information spread across different departments, applications, and platforms. What challenges does this create for product teams, and how do you approach bringing these systems together?
There is no one single approach or solution that will work across the board here, and it largely will depend on the specific enterprise/organization. Their technology architecture and orientation of their engineering teams and services will heavily weigh in on the solutions.
However, one approach I have extensively used is to follow Master Data Management principles, which designates each department as the data owner for the data elements they generate. Depending on the technical expertise and the engineering capabilities they may have, each department, for instance HR, can create a curated data layer that brings the key elements of what their stakeholders data needs (i.e. other departments such as customer support, or sales that may need personnel data).
Each stakeholder group can read from this same curated data layer which ensures all systems are “stitched-together” with the same data. Similarly, the customer support department as the master data owner for the number of support tickets, response and resolution rates etc., should enable a curated data layer for this data. There are other alternatives where some enterprises leverage large data warehousing teams to manage data centrally for the entire enterprise, so like I said there are other alternative solutions and it really depends on the enterprise. At times, the investments into a central data warehouse and global data team provide the business justification to standardize the reporting of common data layers to all stakeholders with appropriate data governance and data security.
Many enterprises do so with large global capability centers based in India. At the end of the day, the product teams need to navigate this enterprise maze and make sure that they are able to identify the right source systems and can directly tap into them for accessing information to align well with the enterprise data.
Can you share an example from your own work where integrating data from different sources was particularly challenging? What decisions did you make, and what was the outcome?
I worked as an AI and data PM for a large team of account managers and they needed access to their customers’ data across various business pillars. As an example, for an online retailer this would be their selection and catalog data, supply chain data, pricing and promotions data, any marketplace fees etc.
For a 360 degree view of the business for their account managers, we needed to curate data and metrics across various different systems. We also needed to understand the right data granularity, and the right definition of the metric to ensure we are using it for the right purpose. From a technical standpoint some decisions and tradeoffs we had to make were to assess if our use cases require access to daily orders data. I had weekly snapshots of the same order data and after discussions with key business stakeholders, I was able to convince them to look at the week over week trends since the primary intent was a monthly/quarterly business review. One of the challenges if we get to process daily orders was that the data size of pour product would increase significantly. This in turn increases application latency and will eventually require higher compute. I associate the data freshness need and the detailed metric requirements specifically based on the underlying use case..
When developing enterprise products, how do you determine whether existing data infrastructure can support what you’re trying to build or whether more substantial changes are needed?
It is definitely a balancing act. If for example, the underlying business need does not warrant the need for certain data and an aggregated metric will suffice, there is no need to overload the underlying infrastructure with additional storage and compute needs.
In reverse, that is exactly the investment needed if latency is beginning to be an issue and you need more optimization. Specifically for healthy product and engineering partnership, a good operating model is to have a robust data dictionary with performance metrics to develop shared understanding for the existing data infrastructure.
Data quality is a persistent challenge for large organizations. How have issues such as inconsistent records, outdated information, or conflicting data definitions affected products you’ve worked on, and how did your team address them?
Missing data is not always bad. Additionally, the business may have evolved and when you start collecting additional information some data might be missing. This is sometimes a byproduct of the fast-paced growth of an organization or an enterprise’s ever-changing needs over time. All of the historical data may not need to be restated and retrospectively filled. Embracing this lapse in data quality is the first thing product teams need to align on.
However, in many cases, the data quality gaps can be corrected, especially if it is publicly available data. For example, while working with life sciences organizations, pharma enterprises can scrub a physician’s “state license numbers” across data platforms. There are industry specific software enterprises that aggregate such publicly available information and offer it as a service. This can enable a pharma company to address any data quality issues in the Physician data generated by their sales teams and their CRM systems for physicians targeted by the pharma company for product sampling and sales calls.
Enterprise products frequently need to access information across systems with different security requirements and ownership structures. How do you approach these constraints when defining product requirements and working with engineering teams?
I prefer to operate under the master data management principles which allows me to combat these ownership structures. For example, HR systems should exert tight controllership and data governance on the data they sell. Their engineering and product teams should exert the underlying data security requirements as they are closest to their data retention and security needs.
This is especially important for financial systems, and enables products and applications through aggregating data for business and operation decisions. From a data governance perspective, only relevant data entities and attributes are made available to application engineering teams and can be codified in the product requirements.
Product leaders are often under pressure to deliver new features while also addressing underlying technical limitations. How do you make the case for investing in data infrastructure, particularly when the benefits may not be immediately visible to customers?
This is where enterprises need to have a long-term view and prepare for the upcoming agentic ere in enterprise data. Moving forward, human-guided and autonomous AI Agents will take over as the primary agent for consuming and creating data. At the same time, your own product’s AI agents needs will be heavily dependent on your own data foundation. Enterprises and product managers that want to lead in the agentic era will tie these use cases in order to rationalize investments in existing data infrastructure and data foundation. There are competing industry opinions about the scale of the growth or the slope of the curve, but most organizations are beginning to embrace themselves for a future with AI agents and agentic workflows for business processes.
Looking at the changes you’ve seen in enterprise technology over your career, how has the role of data architecture evolved in product development? What do you think product managers need to understand about these systems that they might have overlooked five or ten years ago?
Up until now, it was all about humans generating and interacting with the data, but with the onset of AI agents this paradigm will significantly shift in the near future and it can be argued that is already mainstream at some of the leading tech and software services organizations. Product managers need to understand the deep technical details and the AI architecture behind AI applications such as a chat-based AI-agent responding in natural language. This space is also evolving quickly and the patterns that were available a year ago have changed significantly in the last year, so product managers need to stay in lock step with their Engineering teams. This technical depth may have been overlooked five years ago, but now is an essential tenet for leading product managers.

Ricardo Tomaz is VP of Finance & Operations at Ztudium, the 4IR technology company behind Businessabc, Citiesabc, and IntelligentHQ. He owns the group's finance and operations function, spanning financial strategy, budgeting, reporting, and multi-entity operations, and works directly with leadership to align commercial goals with data-driven insight.
Working at the core of a technology group operating across AI and blockchain, Ricardo writes on the practical intersection of finance and emerging technology: how automation and digital transformation are reshaping business operations, what data-driven decision-making looks like in practice, and the questions of governance and ethics that come with AI-led systems.
