A sensible beginner roadmap:
- PL-900 (Power Platform Fundamentals) first. It's not retiring, it's being refreshed in 2026, and it's the right orientation layer for understanding Power Apps, Power Automate, Power BI, and Copilot Studio before you go deeper.
- Then AB-410 as your associate-level target. It reflects where the platform is heading, apps, automation, agents, and AI models working together, which is exactly what employers will be hiring for.
On your other questions:
Resources: Microsoft Learn is the free backbone. What actually builds skill, though, is a free Power Platform developer environment where you build real things. Pick small projects (a leave-request app, an approval flow, a Dataverse-backed app, a Copilot Studio agent over a set of docs) and finish them end to end. For video, creators like Reza Dorrani and Shane Young for Power Apps and April Dunnam for low-code generally are solid. But projects beat playlists every time.
A career without heavy traditional programming: yes, absolutely, that's the whole point of low-code, and functional or maker roles are real careers. The honest angle for you specifically, though: don't drop your software engineering background, lean on it. The most valuable and resilient people here blend low-code with some pro-code (Power Fx, JavaScript, custom connectors, APIs, Dataverse modeling, a bit of Azure). That fusion profile is far more hireable than someone who only clicks through the designer, and your degree gives you a real head start.
Demand: there's steady demand because so many organizations run on Microsoft 365 and Dynamics, and the Copilot and agent push is increasing it. Honestly though, "Power Platform developer" isn't always its own entry-level title. For fresh grads it often shows up inside broader roles like junior developer, business analyst, or Dynamics consultant, and consulting tends to have the most remote-friendly openings. Certs help you get noticed but won't carry you alone, so check current postings in your region to see what they actually ask for.
Skills employers value beyond certs: a portfolio of things you've actually built, the ability to turn a vague business requirement into a working solution, clean Dataverse data modeling, solid ALM (solutions, environments, source control and pipelines), integration sense, and increasingly Copilot Studio and agent building. For functional roles, communication and stakeholder skills matter just as much as the technical ones.
You're starting at a good time with a useful background. Build a few real projects, target PL-900 then AB-410, and keep one foot in the pro-code world. Happy to suggest specific project ideas if that would help.
