Struggling to implement the lex method correctly in my parser

I'm working on a custom language parser and I've been reading a lot about using the lex method for tokenization. My current implementation seems to be failing when I encounter mixed-case identifiers followed by numerical constants. I've tried adjusting the regular expressions, but the parser still throws an unexpected token error. Has anyone here successfully used the lex method with complex grammar rules? I'm specifically struggling with the lookahead functionality when defining boundaries between different data types. Any pointers on best practices or common pitfalls would be greatly appreciated.