
Sign up to save your podcasts
Or


The future of software isn't just generating code—it's knowing what to build and why. We sit down with Cameron Hotchkies, a Staff Software Engineer at Block, focused on data reliability from the lens of traditional site reliability. They decode the hidden mechanics of AI-assisted development. From the meditative focus of woodworking to managing multi-agent AI systems, discover why "vibe coding" won't survive a 3 AM system crash.
We go beyond loose chat to find the hard truths: What happens when "vibe-coded" AI software hits the wall of a 3 AM production crash? And why is "tech debt" actually the ultimate strategic leverage if used correctly?
We explore the critical difference between disposable scripts and load-bearing architecture, the rebirth of the Waterfall method in the age of prompt engineering, and why human intuition remains the ultimate safeguard against algorithmic chaos. Cameron provides the actionable blueprints you need to navigate the rapidly changing landscape of software development and system design.
Connect with Cameron on LinkedIn: https://www.linkedin.com/in/chotchkies/ or at his website: https://semisafe.com
Time Stamps:
(00:00) Intro
(01:08) Analog hobbies: Woodworking and modular synthesizers
(06:29) Rediscovering Waterfall: Research, Plan, and Implement
(15:45) Why AI won't kill software engineering
(18:04) Disposable AI scripts vs. load-bearing architecture
(25:00) AI will not go on-call for you
(41:12) Project estimation and planning with AI
(48:21) Reframing tech debt as engineering leverage
(01:02:24) Career wisdom: Always build in public
Support the pod:
https://3reate.com
https://ko-fi.com/3reate
https://patreon.com/3reate
Listen:
https://podcasts.apple.com/us/podcast/3reate/id1723426314 https://open.spotify.com/show/48Y2M7Ppja43Uq2wlyUtPF https://youtu.be/2wEMD8EvB9I?si=G3iUBE-z4Mx0Ng-Y
By 3reateThe future of software isn't just generating code—it's knowing what to build and why. We sit down with Cameron Hotchkies, a Staff Software Engineer at Block, focused on data reliability from the lens of traditional site reliability. They decode the hidden mechanics of AI-assisted development. From the meditative focus of woodworking to managing multi-agent AI systems, discover why "vibe coding" won't survive a 3 AM system crash.
We go beyond loose chat to find the hard truths: What happens when "vibe-coded" AI software hits the wall of a 3 AM production crash? And why is "tech debt" actually the ultimate strategic leverage if used correctly?
We explore the critical difference between disposable scripts and load-bearing architecture, the rebirth of the Waterfall method in the age of prompt engineering, and why human intuition remains the ultimate safeguard against algorithmic chaos. Cameron provides the actionable blueprints you need to navigate the rapidly changing landscape of software development and system design.
Connect with Cameron on LinkedIn: https://www.linkedin.com/in/chotchkies/ or at his website: https://semisafe.com
Time Stamps:
(00:00) Intro
(01:08) Analog hobbies: Woodworking and modular synthesizers
(06:29) Rediscovering Waterfall: Research, Plan, and Implement
(15:45) Why AI won't kill software engineering
(18:04) Disposable AI scripts vs. load-bearing architecture
(25:00) AI will not go on-call for you
(41:12) Project estimation and planning with AI
(48:21) Reframing tech debt as engineering leverage
(01:02:24) Career wisdom: Always build in public
Support the pod:
https://3reate.com
https://ko-fi.com/3reate
https://patreon.com/3reate
Listen:
https://podcasts.apple.com/us/podcast/3reate/id1723426314 https://open.spotify.com/show/48Y2M7Ppja43Uq2wlyUtPF https://youtu.be/2wEMD8EvB9I?si=G3iUBE-z4Mx0Ng-Y