October 2026 Meetup Minutes

October 24, 2026

Hi,

I have given below the MOM I have captured for the discussion today.

Sessions

Speaker: Kayalvizhi

Discussion Summary

Explored concepts around multi-agent reasoning and collaborative AI systems.
Highlighted that assigning distinct roles to agents does not necessarily lead to diverse reasoning outcomes over time.
Discussed conversation orchestration, state management, communication mechanisms, and agent interactions.
Covered the importance of maintaining agent profiles, tracking conversation progress, evaluating interactions, and managing agent lifecycle events.

Speaker: Rengaraj

Discussion Summary

Presented a simulation-based approach to understanding transistor behaviour.
Discussed the generic working principles of bipolar junction transistors and how simulations can aid learning and experimentation.
Simulation provides a practical way to understand complex engineering concepts.
Software-based experimentation can complement theoretical understanding.

Speaker: Vijay Kumar B

Discussion Summary

Discussed challenges associated with using cloud-based AI models for customer-sensitive data.
Explored the growing importance of local AI infrastructure and high-performance AI workstations.
Reviewed optimisation techniques that improve AI performance, responsiveness, and resource utilisation.
Covered foundational concepts behind how modern language models understand context and generate responses.

Speaker: Praveen

Discussion Summary

Presented the evolution of the developer workflow before and after introducing embedded scripting capabilities.
Discussed the flexibility and productivity improvements that scripting can provide within applications.
Also shared an experimental use case from personal experience involving thermoelectric cooling technology for ligament and tissue recovery.

Speaker: Madav

Discussion Summary

Discussed practical limitations of smaller language models (e.g. 7B), including accuracy and context retention challenges.
Highlighted the inherently probabilistic nature of AI systems and the resulting variability in outputs.
Explored collaborative AI approaches where different models perform complementary functions.
Presented “Mixture of Agents” concepts for improving solution quality through multiple perspectives.

Speaker: Sakthi

Discussion Summary

Encouraged teams to actively experiment with emerging AI technologies and approaches.
Introduced the SystemOne model and discussed potential application areas.
Called for experimentation using Jev for a SystemOne based request processing.

Thanks,
GuruVinayak P