I just recertified one of my AWS professional credentials without sitting the exam again. Not because I skipped anything — because AWS changed the pathway, and the change is worth paying attention to.
It's called Maintain, and it replaces the old model — cram for months, sit a high-stakes exam, repeat every few years — with something closer to how skills actually stay current in the field. Instead of a single retake, professional-level architects can extend their certification by earning 700+ points through courses and hands-on labs, and completing at least two practical, scenario-based activities. Hit the bar, and the credential extends through 2027.
I finished at 760 points, with both practicals done. But the number that matters isn't mine — it's what AWS chose to put in front of architects to earn those points with.
My actual Maintain progress on AWS Skill Builder — 760/700 points, both practicals complete.
The catalog is the signal
When a certification body redesigns how you stay current, what they put in the course catalog tells you what they think the role is turning into. This one leans hard into a specific direction: Security and Governance for AI Solutions, Automating GenAI Workflows with Amazon Bedrock Flows, Building Agentic Applications the Well-Architected Way.
The actual activity list — the AI and agentic courses aren't buried, they're front and center.
That's not incidental. AWS builds these programs around where they expect real production workloads to go next. A recertification pathway is, among other things, a fairly reliable leading indicator — and right now it's pointing straight at AI and agentic systems as core architecture work, not an optional specialty bolted onto the "real" job.
Three things worth taking from this
- Continuous learning is starting to outperform exam cramming, by design. A pathway built around ongoing hands-on work rewards people who stay engaged with the platform year-round, not people who binge-study for six weeks every three years. That's a better incentive structure, and it's also a harder one to fake.
- The hands-on labs are the actual value, not the badge. AWS's SimuLearn-style practical activities put you inside a real, scenario-based environment rather than a multiple-choice exam simulator. The skill you walk away with is closer to what the job actually requires than what a traditional exam ever tested.
- Agentic AI knowledge just moved from "nice to have" to "core architect skill." If you're a cloud architect and your AI exposure stops at "I've used ChatGPT a few times," the gap between where the certification body is pointing and where your actual skills sit is wider than it looks from the outside.
Why this matters beyond one credential
I've spent two decades moving through AWS, Intel, IBM, and Citrix, and the pattern repeats every time a platform matures: the people who stay ahead aren't the ones who pass a test once and stop. They're the ones who treat the underlying skill as something to keep building, quietly, in the background, whether or not a recertification deadline is forcing the issue.
That's the same principle we build NovuSpark's own programs around — not a certificate of attendance, but a skill that's still being used, and still growing, long after the session ends.