AI Agents, Intent, and Testing: What Engineering Looks Like Now with Leonid Igolnik
Failed to add items
Add to cart failed.
Add to wishlist failed.
Remove from wishlist failed.
Follow podcast failed
Unfollow podcast failed
-
Narrated by:
-
Written by:
About this listen
Firas Sozan sits down with Leonid Igolnik, Chief Technology Officer at Clari, to unpack what AI agents are, and how AI is changing software engineering, hiring, and the fundamentals of building reliable systems. You will learn why intent expression matters, why testing is becoming the guardrail for non-deterministic outputs, and how agents will depend on context, not just data.
Episode Description: AI is a new level of abstraction, but the hype is running ahead of clarity. In this conversation, we break down what that means for engineers, leaders, and founders trying to build and operate real software in a world where AI writes more code faster, without automatically improving quality or security.
We explore how interview loops need to evolve when candidates have full access to AI tools, and why mastery of fundamentals still matters. The key shift is intent, the ability to clearly express what you want, validate results, and operationalize systems reliably, especially in enterprise environments where determinism and trust are non-negotiable.
We also dig into the agent wave, why everyone wants an agent, why definitions are still blurry, and why context is the unlock for real value. Finally, we zoom out to the business side, and discuss how AI could amplify revenue teams without replacing the human function entirely.
We cover: • How AI changes the definition of engineering skill and evaluation • Why testing and code coverage become practical guardrails • What “intent” means when AI generates the first draft of code • Python default behavior in modern AI coding tools and what it signals • Why runtime maturity and operations matter beyond language choice • What agents can automate in revenue workflows, and what cannot • Why one engineer startups fail at scale, even with AI leverage
Who this is for: Founders, operators, and engineering leaders who want a practical model for using AI tools without losing quality, security, or execution discipline.
Links and Resources: Spotify: https://bit.ly/spotify-itsm Apple Podcasts: https://bit.ly/apple-itsm Amazon Music: https://bit.ly/amazon-itsm Website: https://insidethesiliconmind.com/ Follow on LinkedIn: https://www.linkedin.com/company/inside-the-silicon-mind/ Instagram: https://www.instagram.com/insidethesiliconmind/ TikTok: https://www.tiktok.com/@insidethesiliconmind X: https://x.com/InSiliconMind
Mentioned in This Episode: Made to Stick Thinking, Fast and Slow
About the Show: Inside the Silicon Mind is your masterclass in high-stakes innovation, business strategy, and the Silicon Valley mindset. Hosted by Firas Sozan, we interview Founders, CEOs, and Venture Capitalists shaping the future of technology.