What happens when technical brilliance meets real-world impact? You get Abhishek Dodda. He has over a decade at Visa and a background spanning Nike, Bank of America, and The Home Depot. Abhishek has shaped the architecture of data and AI at an enterprise scale.
In this interview, he opens up about building scalable Gen AI platforms, pushing blockchain innovation, and leading product teams that deliver with purpose. Abhishek knows how to make tech work for businesses that move billions. From launching AI-powered APIs to reducing disputes and streamlining infrastructure, Abhishek talks about what it really takes to build future-ready systems. This interview is an unfiltered dive into what drives real progress in tech.
Q1: Abhishek, thank you for joining us. You have had such an impressive journey through data engineering roles at institutions like Bank of America and leading Gen AI initiatives at Visa. Can you walk us through the mindset shifts that guided your transition from technical execution to strategic product leadership in data and AI?
Abhishek Dodda: Absolutely. My journey began with a deep focus on technical execution—designing data architectures, implementing analytics pipelines, and deploying AI models. But as I moved through roles at institutions like Bank of America and later at Visa, I realized the importance of aligning technology with business outcomes.
The key mindset shift was recognizing that technology isn’t just a support function—it’s a strategic driver of growth and innovation. I began to focus more on understanding stakeholder needs, long-term organizational goals, and market trends. This shift required building cross-functional collaboration skills, adopting a product-oriented thinking approach, and staying grounded in measurable impact. It was no longer about “what can the tech do?” but rather “how does this move the needle for the business and its customers?”
Q2: In your role as Senior Manager of B2B Payments at Visa, you’ve led initiatives like the Enterprise RAG and Model as a Service for scalable AI solutions. What were the biggest technical and organizational hurdles in deploying Gen AI at scale, and how did you overcome them?
Abhishek Dodda: One major technical challenge was ensuring the scalability and security of generative AI models across Visa’s complex, global infrastructure. Building out the Enterprise Retrieval-Augmented Generation (RAG) framework and Model-as-a-Service required not only a robust architecture but also strong data governance and privacy safeguards.
Organizationally, there was the challenge of stakeholder alignment and change management. We overcame this by creating cross-functional task forces and clear communication channels. I also invested heavily in educating business units on AI capabilities and limitations. By focusing on quick wins—like reducing transaction disputes by 25% through AI—we demonstrated value early and built trust for wider adoption.
Q3: You’ve successfully applied blockchain and generative AI for operations and reduced disputes while also leading large-scale data projects at major companies like Nike, Home Depot, and Bank of America. Can you describe a specific initiative where these technologies intersected with business strategy to deliver measurable impact, such as cost savings, job creation, or product innovation?
Abhishek Dodda: Certainly. One notable initiative at Visa involved integrating generative AI with blockchain-based financial applications to streamline and secure dispute resolution processes. By using AI to analyze transaction patterns and customer behavior, we could preemptively detect anomalies, while blockchain provided an immutable trail for verification.
This dual-tech approach led to a 25% reduction in disputes, enhanced transparency for both banks and merchants, and reduced operational overhead. The initiative directly supported Visa’s strategic goals of enhancing customer trust and reducing friction in payment systems, while also saving millions in potential dispute-related costs.
Q4: Your work speaks volumes about data’s potential to drive tangible business outcomes. What is your approach to ensuring that advanced analytics remain tightly aligned with end-user value and trust?
Abhishek Dodda: I follow a principle-driven approach that balances innovation with integrity. First, I ensure that all analytics initiatives begin with a clear understanding of the end-user problem and measurable success criteria. I then embed explainability and fairness into model design, making outputs interpretable and actionable.
Equally important is continuous stakeholder engagement—getting feedback early and often. I also advocate for transparent data policies and ethical AI frameworks that reinforce trust. Ultimately, if users don’t trust the outcomes, the analytics won’t drive real adoption, no matter how advanced they are.
Q5: In your Visa experience, you reference piloting Groq’s Mixture-of-Agents and integrating semantic caching. These are the newest technologies. How do you evaluate emerging Gen AI tools for enterprise readiness and ensure their ethical and cost-effective implementation?
Abhishek Dodda: Evaluation starts with a rigorous technical assessment—scalability, integration potential, and alignment with internal security and compliance standards. In piloting tools like Groq’s Mixture-of-Agents and semantic caching, we conducted sandbox testing with real data to validate performance and value.
Ethical implementation is non-negotiable. I involve cross-disciplinary teams, including legal, compliance, and HR, to ensure new tools meet ethical guidelines and data privacy laws. And for cost-effectiveness, we always benchmark performance against ROI metrics. If a tool doesn’t show scalable value, we pivot quickly. It’s about balancing innovation with accountability.
Q6: Your leadership of global teams across the US, Singapore, and India, especially during your time modernizing Visa’s data platform, must have required strong cultural and operational dexterity. How do you ensure innovation and cohesion in globally distributed engineering teams?
Abhishek Dodda: Leading teams across the US, Singapore, and India, especially during the modernization of Visa’s data platform, required intentional culture-building. I prioritized three things: clarity, consistency, and connection.
We established a unified mission and success metrics so every team understood their role in the broader vision. We also standardized tools and processes to eliminate friction. But most importantly, I created opportunities for shared innovation—hackathons, knowledge-sharing sessions, and rotational leadership programs.
Cultural sensitivity and timezone respect were essential. I made it a point to visit teams physically where possible and maintained an open-door policy virtually. When teams feel valued and connected, innovation flourishes—no matter where they are.
Conclusion
Abhishek Dodda emphasizes and materializes building platforms that scale, last, and evolve. He gives a clear sense of how strategy, leadership, and technical depth intersect in his work. Whether it’s Gen AI frameworks or global B2B payment systems, his approach combines experimentation with focus. His mindset, grounded in staying curious and always pushing boundaries without chasing buzzwords, is what drives him to success.
This success isn’t luck; it’s the result of taking tough challenges and turning them into structured, measurable outcomes. He has gone from hackathons to enterprise deployments, reflecting a thoughtful and bold journey. Abhishek knows that the future of AI and blockchain isn’t just about innovation. It’s about knowing how to make it real, and he’s successfully fulfilling this mission.