Outshift by Cisco published a whitepaper titled "Scaling out Superintelligence" in January that suggests that new standard network layers will be required for AI agents and humans to work together at scale. It also began designing a distributed infrastructure to support those layers, which it calls the "Internet of Cognition," that helps agents and humans think together with a shared semantic context. In this episode, Guillaume de Saint Marc, vice president of engineering at Outshift, discusses how the project evolved and what's coming next.

Featuring: Guillaume de Saint Marc, VP of Engineering, Outshift by Cisco

In today’s episode, we’ll cover…

• Artificial superintelligence – definition and challenges

• Proposed OSI model Layers 8 and 9 for agentic AI

• The Internet of Cognition vs Agntcy and A2A protocols

And more!

References:

• Big vendors back Linux Foundation agentic workflows project

• How agentic AI is changing work, strategy and competitiveness

• Is your business ready for an agentic AI team?

• The ethics that make human-AI agent collaboration work

To learn more about software development and platform engineering, check out Search IT Operations.

To watch clips from our podcast, subscribe to our YouTube channel, @EyeOnTech.

There are two sides to the mainframe modernization coin, according to Asa Kalavade, vice president of AWS Transform, an agentic AI service that modernizes Windows, mainframe, and VMware environments. In late February, Asa wrote a blog post about misconceptions regarding AI and mainframe modernization that were revealed by work with more than 400 AWS Transform customers. In that post and during this episode's discussion, she explains why successful mainframe modernization requires both reverse and forward engineering – and why AI is only good at one of those things.

Featuring: Asa Kalavade, VP of AWS Transform

In today’s episode, we’ll cover…

• The need for both reverse and forward engineering in app modernization
• The intricacies of mainframe applications AI tends to struggle with
• Where AI agents help the modernization process
• The evolving role of mainframe experts

And more!

References:

• How to use ChatGPT in mainframe modernization
• Mainframe modernization to AI apps: USPTO reinvents itself
• Learnings from COBOL modernization in the real world (AWS)

To learn more about software development and platform engineering, check out Search IT Operations.

To watch clips from our podcast, subscribe to our YouTube channel, @EyeOnTech.

"I would love there to be a SaaS apocalypse" in observability – that's the spicy take from Ari Zilka, CEO at data pipeline engine startup MyDecisive.ai, and a former exec at New Relic and Hortonworks. MyDecisive's open source software targets data management costs from observability vendors such as Datadog by reducing data at the source; it can also proactively alert on and take action on OpenTelemetry data as it comes in from enterprise applications. If Zilka has his way, observability will return to enterprises as an in-house discipline, but without requiring an army of engineers to set it up.

Featuring: Ari Zilka, CEO at MyDecisive.ai

In today’s episode, we’ll cover…

• MyDecisive's approach to AI and Zilka's outlook on AI agents (It's not what you might think)
• Observability costs and how MyDecisive proposes tackling them
• MyDecisive vs Cribl and other data pipeline tools

And more!

References:

• New Relic plans to expand AI agent observability
• Snowflake storms into IT monitoring with Observe acquisition
• Dynatrace AI agents draw on new observability integrations

To learn more about software development and platform engineering, check out Search IT Operations.

To watch clips from our podcast, subscribe to our YouTube channel, @EyeOnTech.