Revisiting the SDLC: Waterfall, Agile, and the Lessons That Still Hold

One of the first things I revisited while studying Managing AI Application Development was the Software Development Life Cycle, specifically the traditional Waterfall model. Waterfall lays everything out in a straight line: requirements, design, build, test, deploy, maintain. Each phase finishes before the next one starts. On paper it is clean and predictable, which is exactly why it was the default approach for decades, especially in large, compliance-heavy organizations where documentation and sign-off matter as much as the software itself.

But sitting with Waterfall again, I kept thinking about how different my actual working experience has been. At Volanno, supporting a proprietary government web application for the FAA, we operated in an Agile environment with weekly sprints, sprint retrospectives, and planning poker. That environment is where I earned my PSM1 (Professional Scrum Master I) certification from Scrum.org in 2019. Revisiting the theory behind Waterfall made me appreciate just how much Agile changed the way I think about shipping software.

Waterfall assumes you know everything upfront. Agile assumes you don't, and builds in the feedback loops to adjust as you learn. Weekly sprints forced our team to show working progress constantly rather than waiting months for a single, high-stakes deployment. Sprint retrospectives gave us a structured moment to ask what actually worked and what didn't, instead of burying those lessons until a post-mortem after launch. Planning poker turned estimation into a team conversation instead of a guess made in isolation.

Going through the SDLC material again as part of this Quantic course is a good reminder that neither model is inherently right or wrong. Waterfall still has a place in projects with fixed, well-understood requirements and heavy regulatory constraints. But for AI application development specifically, where the requirements and even the capabilities of the models themselves can shift mid-project, the Agile mindset I built at Volanno feels more relevant than ever. You're not just building software, you're building something that has to adapt as the underlying technology evolves.

Attached below are my completion certificates for Managing AI Application Development I and II from Quantic. More checkpoints to come as I work through the rest of the AI Engineering program.

This reflects my personal learning experience and does not represent Quantic or reproduce course materials