After sequencing 192 clones I came to realize that a couple dozen of my reads have clusters of N's at the beginning and end of the contigs. Are there any good references or rules of thumb for where to cut off portions with N's? For example, if a N shows up on my read on the 12th base, should I cut off all 12 bases? Cheers.

You should look at the quality of the chromatogram for each sequence file to determine why that residue was classified as an N.

If it is just that base that is misread then you don't have to eliminate the surrounding sequence. If all the read is poor quality in that region remove the whole region until the quality is sufficient to be confident of the data.

Don't blindly accept automated sequencing results. Always manually inspect the quality of the chromatograms you receive before accepting the data.