I've been away from serious OTB chess for quite a while now (though we do have a club at work where I still play some casual games) and so most of my chess activity these days is online, where I'm playing blitz or correspondence (but I'm really trying to avoid the former!). I played a couple of rapid (30 min) games online recently, so here they are for your viewing pleasure. I've been playing the French quite a bit, though I don't know all the theory. I looked over the games a bit with Stockfish.
When I saw the knight was trapped, I immediately played b5 as I didn't see that c4 was a better way to continue, gaining space, protecting the pawn and still leaving the knight trapped. 12. ... Bb7: the choice really was between this and f6, but I decided upon Bb7 to continue with development (probably due to being taught to develop pieces and still not recognising when it might be better to do something else). Perhaps I could have played f6 as early as move 7, instead of a6 and c5. 16: ... Ndxe5: this is not a move I'd normally play, but it just looked kinda fun. Just a normal move like Rc8 would have been better. I missed d4 on move 22, too, giving the bishop a bit more of a diagonal and then aiming to go Rc4 on Qxd4.
I ended up resigning, as I felt that I must be worse. It felt very uncomfortable to defend.
First of all, I played 5. ... Bg4 to exploit White playing Bd3. However, by move 14 this bishop looks out of place

. I suppose why Bd6 is preferred here is just to stop White's bishop coming to f4. Obviously I ended up playing Bd6 anyway, but I guess I ended up wasting time. I spent a whole 5 minutes or so on move 11 and
as you can see, I got into severe time trouble later so really had no time to think when I needed it. I completely missed the continued threat of Ne4 and Ng6 and then didn't see Qxd6 on move 31. Stockfish suggests the ending is better for Black, but even if I had plenty of time, I'd likely not have held it.