In our insightful conversation with Dr. Harshad Khadilkar, an experienced researcher, we explore the wide impact of generative AI. Dr. Khadilkar’s expertise across air and rail transport, energy, and supply chain management enriches our discussion. We will explore the intersection of AI, operations research, and finance. Let’s discover trends in generative AI, learn from Dr. Khadilkar’s career, and see how he envisions using technology to improve finance decisions.

You can listen to this episode of Leading with Data on popular platforms like Spotify, Google Podcasts, and Apple. Pick your favorite to enjoy the insightful content!

Key Insights from our Conversation with Harshad Khadilkar

• Generative AI will undergo rationalization, with its use becoming more specialized and impactful in applicable domains.
• The democratization of generative AI technology is on the horizon, with a focus on creating smaller, cost-effective models.
• India’s research environment has evolved to match global standards, offering significant opportunities for high-end research work.
• Reinforcement learning and generative AI are experiencing a convergence, leading to innovative approaches in AI-driven decision-making.
• Domain expertise is crucial for a successful career in AI; specialization can set candidates apart in a field crowded with generalists.
• Generative AI will serve as a tool to augment human capabilities in finance, not replace them, enhancing efficiency and decision-making processes.
• The future of AI in finance is promising, with potential for significant advancements in portfolio management and investment strategies.

Join our upcoming Leading with Data sessions for insightful discussions with AI and Data Science leaders!

Now, let’s look at the details of our conversation with Harshad Khadilkar!

How do you foresee the evolution of generative AI in the coming years?

As a Senior Scientist at TCS Research, I’ve observed three key trends that are likely to shape the future of generative AI. Firstly, there’s a current exploration of generative AI’s potential across various domains, though it’s clear that it won’t be suitable for every application. Secondly, in areas where generative AI is applicable, we can expect to see its transformative power unleashed. Lastly, there’s a push towards democratizing generative AI technology, aiming for smaller, more affordable models that deliver specialized, high-quality outputs without the hefty price tag.

As you reflect on your career journey, what pivotal decisions did you make?

When I was completing my PhD, I faced the choice of staying in the US or moving to India. I decided to return to India in 2013, sensing the burgeoning opportunities for high-end research work. My career moves since then, from IBM to TCS and now Franklin Templeton, have been driven by my pursuit of impactful research that bridges the gap between theory and real-world application.

How does the research environment in India compare to that in the US?

The landscape has shifted significantly. While core research was traditionally conducted in the US or Europe, with peripheral tasks outsourced to India, companies have recognized the value of India’s engineering talent and cost-effectiveness. Today, major corporations have established substantial research operations in India, working on par with their global counterparts.

Can you share an example of how you approach problem-solving in your research?

At TCS, I tackled the challenge of supply chain inventory management for perishable goods. We developed scalable reinforcement learning algorithms to optimize inventory levels, considering factors like shelf life and seasonal variations. This project was particularly satisfying as it translated into real-world benefits for our clients.

How has the generative AI era impacted your work in reinforcement learning?

The rapid advancements in generative AI, particularly with models like GPT-3.5 and 4, were somewhat unexpected. These models have shown surprising effectiveness in decision-making tasks, revealing fundamental similarities in how they represent language and how reinforcement learning agents represent states. This has opened up new avenues for cross-pollination between fields.

What advice would you give to someone starting a career at the intersection of domain expertise and AI?

Focus on becoming a domain expert first. With many generalists in AI and data science, the demand is for individuals who can apply AI to solve specific domain problems. Building unique expertise in a particular area will provide a strong foothold in the industry.

Looking ahead, what are your professional aspirations for the next few years?

I’m excited to delve deeper into the domain of finance, identifying and addressing the open problems that lie at the intersection of technology and mathematical challenges. My goal is to contribute to solving these issues, leveraging AI to enhance financial decision-making and portfolio management.

How do you envision the role of generative AI in finance and other domains?

Generative AI will affect many areas, but it won’t make a bad future where AI controls everything. Instead, it will be a strong tool that helps people, especially in tasks like analyzing data, spotting patterns, and finding chances.

Summing Up

Dr. Harshad Khadilkar’s perspective on the future of generative AI highlights its transformative potential across diverse sectors. He envisions a landscape where this technology enhances human capabilities, rather than overshadowing them entirely. He advises newcomers in the field of AI on how to position themselves for success in an increasingly competitive environment.

For more engaging sessions on AI, data science, and GenAI, stay tuned with us on Leading with Data.

Check our upcoming sessions here.