While building the Bangla Typing platform, the most intricate technical hurdle was grapheme cluster segmentation in Unicode.
Unlike Latin scripts, a single conceptual Bengali character like "ক্ষ" is composed of multiple code points (Ka + Virama + Ssa). Standard JavaScript string indexing treats these as multiple characters.
If not handled properly, backspacing, cursor position, and error penalty calculations fail. Using modern `Intl.Segmenter` and Unicode normalization (NFC) ensured precise keystroke verification across all 13 curriculum levels.