16 Aug 2025
46m

Lessons Learned from Building Agentic Systems With Jayeeta Putatunda

Podcast cover

AI Explained

This webinar episode of AI Explained features Krishna Gadhe, CEO of Fiddler AI, interviewing Jayeeta Putatunda, Director of AI Center of Excellence at Fitch Group, about lessons learned from building agentic systems. Jayeeta discusses the importance of aligning AI projects with business value, using the 80-20 rule to focus on high-impact use cases, and establishing specific metrics for evaluating generative AI systems. She differentiates between workflow-oriented systems and fully autonomous agents, emphasizing the need for credibility, responsibility, and accountability in AI applications, particularly in finance. The conversation covers testing methodologies for non-deterministic systems, the role of observability, and strategies for mitigating hallucinations in financial contexts by integrating classical machine learning models. Jayeeta also shares insights on bridging the gap between agentic system capabilities and enterprise needs, highlighting the significance of collaboration between technical teams and business stakeholders.

Outlines

Part 1: Introduction and Foundations

Part 2: Testing, Evaluation, and Observability

Part 3: Challenges and Solutions

Part 4: Development and Best Practices

Part 5: Use Cases and Trust

Sign in to continue reading, translating and more.

Open full episode in Podwise