Currently working on Tenstorrent bounties, focused on low-level performance optimisation.
- Accurate sin/cos/tan on Tenstorrent
- Cube Root on Tenstorrent
- In-Place 32ร32 Matrix Transpose on Tenstorrent
- 32-bit Integer Max/Min on Tenstorrent
- 16-bit Integer Multiplication on Tenstorrent
- Typecast on Tenstorrent
- Rounding on Tenstorrent
- 32-bit Integer Division on Tenstorrent
- 32-bit Integer Multiplication on Tenstorrent
- Optimal "where" on Tenstorrent





