LLM Observability and LLM Security

** Please note that, this event and the sessions are curated for folks with 2+ years of working experience **

Agenda:

1. 9:45 AM - Registration
2. 10:00 AM – 10:15 AM - Ice Breaker Activity By: Blessing Joshwa
3. 10:15 AM – 10:20 AM - Welcome Note, By: Dharan K T
4. 10:20 AM – 11:20 AM (60 mins) - LLM Observability by Sivasubramian Bagavathiappan, Co-Founder, GuhaTek - https://www.linkedin.com/in/kbsivacse/
5. 11:20 AM – 11:40 AM - Networking Break & Refreshments
6. 11.40 AM to 12.00 PM (20 mins) - Lightning talk 1
7. 12.00 PM to 12.20 PM (20 mins) - Lightning talk 2
8. 12:20 PM – 01:20 PM (60 mins) - LLM Security by Madhan Babu Bhoopal, Tech Lead at Bosch Global Software Technologies Ltd - https://www.linkedin.com/in/madhan-babu-bhoopal-b1bb924/

LLM Observability
- As large language models evolve into autonomous agents, understanding their internal behavior becomes critical for reliability, safety, and performance.
- This talk explores emerging techniques for LLM and AI agent. observability, tracing prompts, reasoning paths, and system interactions to uncover how and why models make decisions.
- Attendees will learn practical strategies for instrumenting, visualizing, and debugging complex LLM-powered systems in production.

LLM Security
- As enterprises increasingly integrate Generative AI into their internal ecosystems, securing access to sensitive data becomes a mission-critical challenge.
- This session explores the evolving security landscape of GenAI applications that interact with corporate data through Model Context Protocol (MCP) servers.