Moving from defense ML/signal processing to commercial after 20 years

I've spent the last 20 years doing ML and signal processing research in the defense sector. While I've been fortunate enough to publish (which I know isn't always possible in defense), I'm increasingly frustrated by chronic underfunding and the feeling of always playing catch-up with commercial tech. Some context:

-PhD-level researcher with focus on ML/signal processing

-Multiple publications in IEEE journals

-Extensive experience with radar, sonar, and multiple imaging systems (including MWIR, NIR, DTV, Video, SAS, SAR).

-Led multiple DoD-funded research programs

-Comfortable with both theoretical and applied work

The funding cycles and bureaucracy in DoD work are wearing me down. I see commercial companies moving much faster, with better resources and more modern tech stacks.

For those who've made this transition:

-How different is the research culture in commercial ML?

-What was the biggest adjustment?

-Did you find your defense background valued or did you have to "prove yourself" again?

-Any tips for positioning defense experience for commercial roles?

-Areas where defense experience is particularly valuable?

Would appreciate insights from both sides of this transition.