
Episode #41
Why AI Won't Fix Your Company's Problems — It Will Amplify Them | Rob Broadhead
Good architecture rarely makes headlines, but bad architecture eventually becomes impossible to ignore. Behind every app and dashboard sits a set of decisions about infrastructure, data flow, and scalability — and when those decisions are made poorly, organizations spend years fixing problems built into the foundation. Rob Broadhead is a veteran software architect, entrepreneur, and founder of RB Consulting. Over 30 years, he has designed enterprise systems, advised businesses on technology strategy, and helped organizations avoid costly mistakes from poorly planned technology decisions. He also hosts the Devpreneur Podcast. In this episode, Collins Victory Odabi sits down with Rob to explore the most common architectural mistakes companies make, how to turn technology from a cost center into a genuine growth driver, and why AI amplifies whatever problems already exist in a company rather than solving them. In This Episode, We Discuss The most common architectural mistake is building only for today, not for growth. Rob describes companies painting themselves into a corner with systems that work perfectly at launch but require costly rebuilds once growth arrives faster than expected — the fix is asking "what happens if this grows 10x" at the design stage, not after. Pressure-testing assumptions matters more than accepting the first answer. Rob's example: asking a client if only one user will ever access the system, getting a confident "yes," then finding out six months later that families or organizations need multiple logins — a gap that better upfront questioning would have caught. Modernizing legacy systems means more than a straight lift-and-shift. Rob argues that migrating an old system into a modern stack without also adopting new capabilities — mobile access, modern interfaces — wastes the investment; if you're already spending the time and money, take full advantage of what modern technology actually offers. Fundamentals matter more than specific programming languages for developers building long-term careers. Rob has shifted his mentoring advice toward architecture, software development lifecycle, and problem definition, since knowing a particular language is becoming less valuable as AI increasingly handles implementation itself. AI codes like a junior developer — technically capable, but incapable of system-level thinking. Rob argues that without strong guardrails, clear requirements, and human code review, AI-assisted development produces working code littered with redundancy and technical debt, since AI stops at the first brute-force solution rather than generalizing and refining it. About Rob Broadhead Rob Broadhead is a veteran software architect, entrepreneur, and founder of RB Consulting, and host of the Devpreneur Podcast (Building Better Developers). rb-sns.com | rob@sns.com️ develpreneur.com Key Timestamps / Chapter Markers [0:04] Introduction and episode framing[2:03] From coding in grade school to founding RB Consulting in 2001[6:32] The most common architectural mistakes organizations make[8:53] Aligning technology decisions with long-term business objectives[13:35] Turning technology from a cost center into a growth driver[15:54] Where to start when modernizing legacy technology infrastructure[19:14] What separates developers who thrive long-term from those who struggle[26:50] Why system design matters even more in the age of AI-assisted coding[30:47] The principle behind building systems meant to last for years Connect with Collins LinkedIn: linkedin.com/in/collins-odabi-266620380 Guest enquiries: podmatch.com/member/web3unfiltered Until next time, stay sharp, stay curious, and keep building.

