I'm trying to alkylate a secondary amide carbonyl with a primary grignard compound, but I have a problem. Grignard acts as a base instead of a nucleophile. Any idea to increase nucleophility of grignard and/or avoid the deprotonation in the alfa carbon of the amide?

The product its an N-alkylated pyperidone . I'm trying to introduce an alkyl chain (with a grignard reagent) into de carbonil position, generating the corresponding enamine (cause although the amino could act as a leaving group the resulting amino-ketone after the work-up could be cyclizated again with a Dean-Stark).
But in that case I just obtain an intermolecular by-product that could only be explained if the grignard has acted as a base in the alfa-carbon
As the pyperidone is N-alkylated I can't prepare a Weinreb-kind amide

There are ways to increase the nucleophilicity of anion (transmetalation to Cerium come to mind) but I'm pretty sure that the amide proton is to acidic for that type of strategy to work (usually it is used to add to highly enolizable carbonyls such as tetralone). You will have to protect the amide (depending on your other functional groups a BOC , or Troc carbamate is pretty standard) and then add your nucleophile, and then deprotect.


Cisco

What?, As I told you the lactam is N-alkylated. There isn't any N-H !!!, BOC protection?, where? the acid proton is in the alfa-carbon. If it was an NH I'm not so stupid to know that grignard would act as a base for sure.