Daiteq S.R.O.

daiteq provides advanced arithmetic solutions for space-grade processors and FPGAs. For floating-point processing we offer a highly configurable floating-point units compliant with IEEE 754 (2019) that support complex arithmetic and user-defined floating-point number formats, beneficial e.g. for deep learning. For fixed-point processing we offer an implementation of SIMD operations targeted at GNSS processing, low-precision deep learning inference and video compression. Our arithmetic units are complemented by our customised version of the LLVM compiler that supports user-defined floating-point and integer data types.