In this episode, we’re diving into how orgs can leverage AI to create more autonomous processes when it comes to product development and quality control with Iccha Sethi, SVP of Engineering @ Vanta. Iccha shares how as an eng leader, it’s imperative to continue to be hands on, and she shares her experience replicating Vanta by vibe coding and what it taught her. She not only breaks down Vanta’s product development workflow and the ways in which AI has automated it, but she also shares the guardrails & checkpoints in place that ensure quality. Additionally, Jerry and Iccha discuss individual and team accountability within Vanta’s fast product cycle. Lastly, they cover the AI DX team, who it is comprised of, and its role within Vanta while still maintaining a culture of experimentation when it comes to AI. 

ABOUT ICCHA SETHI

Iccha Sethi is SVP of Engineering at Vanta, where she leads the engineering organization building products that help businesses earn and prove trust by automating security and compliance. Previously, she led engineering teams at GitHub across products including Actions, Codespaces, npm, Packages, and Pages. Her career also includes engineering and leadership roles at InVision, Atlassian, and Rackspace.

Iccha focuses on connecting engineering decisions to business outcomes, building reliable products at scale, and exploring how AI and agentic systems are reshaping software products and engineering organizations. She brings the perspective of both a hands-on technical leader and an executive leading engineering organizations. Outside work, she enjoys reading science fiction and fantasy.

The CAFE(S) framework is a research-backed definition of context quality for AI agents.

DX collaborated with researchers at Capital One, Google, and UVic to give teams a shared standard for reviewing agent context by breaking it into five properties: clarity, actionability, fidelity, efficiency, and security.

Your agent is only as good as its context and the CAFE(S) framework establishes a shared language around context to help you better understand how & why agents go wrong.

Check it out here!

SHOW NOTES:

• Introducing Iccha & her leadership philosophy / style (1:23)
• Differentiating between what is marketing vs. real value for engineers (3:33)
• Lessons learned from vibe cloning Vanta (6:33)
• Iccha’s perspective on the future of SaaS (9:04)
• Understanding total cost of ownership when it comes to app maintenance (10:28)
• How to stay hands on as an eng leader / knowing what to practice (11:48)
• Vanta’s process from customer call through production (14:43)
• Implementing an EPD triad to iterate quickly & allow autonomy (17:48)
• What “Walk the Store” is / how it acts as a checkpoint for the team (19:34)
• Strategies for a more automated product workflow & quality / eng guardrails (22:41)
• What handoffs look like @ Vanta (25:22)
• Creating a culture of accountability within Vanta’s product cycles (28:12)
• Frameworks for scaling agent-assisted workflows (30:44)
• Forming an AI DX team & its overall goals (32:59)
• Qualities that make a successful AI DX team member (35:21)
• Supporting bottom-up experimentation without creating internal chaos (37:55)
• Rapid fire questions (40:26)

This episode wouldn’t have been possible without the help of our incredible production team:

Jerry Li - Host

Noah Olberding - Associate Producer, Audio & Video Editor https://www.linkedin.com/in/noah-olberding/

Dan Overheim - Audio Engineer, Dan’s also an avid 3D printer - https://www.bnd3d.com/

Ellie Coggins Angus - Copywriter, Check out her other work at https://elliecoggins.com/about/