Update README.md

README.md

@@ -7,24 +7,19 @@ sdk: gradio
 pinned: false
 ---
 
-Tenstorrent is a next-generation computing company in AI accelerator and RISC-V CPU designs. 
-Tenstorrent is bringing to market highly programmable chips at an accessible price point as a means of driving innovation. Tenstorrent brings together experts in the field of computer architecture, ASIC design, RISC-V technology, advanced systems, and neural network compilers.
+Tenstorrent is a next-generation computing company building AI accelerator and RISC-V CPU designs. Tenstorrent is bringing to market highly programmable chips at an accessible price point as a means of driving innovation. 
+Tenstorrent has developed a proprietary processor core called the Tensix which it uses to make its own AI chips, but also sells its IP and other technology to customers looking to make their own AI chips.
 
-Tenstorrent makes its own AI chips, but also sells its intellectual property and other technology to customers looking to make their own AI chips.
-Tenstorrent has developed a proprietary processor core called the Tensix; released two products, Grayskull (machine learning processor) and Wormhole (networked machine learning processor); and is developing the industry's first 8-wide decoding RISC-V core capable of addressing both client and HPC workloads.
-
-Tenstorrent has opened source its software stacks:
+Tenstorrent has two open source software stacks:
 TT-Buda (https://github.com/tenstorrent/tt-buda), and 
 TT-Metalium (https://github.com/tenstorrent-metal/tt-metal/)
 
-
 To contact us, please reach to our Discord channel (https://discord.gg/tenstorrent)
 
 -------
 
-TT-Metalium’s bare-metal programming stack defines a low-level programming model to allow users to write custom kernels for deployment on Tenstorrent hardware. It comes with debug and performance profiling tools. 
-
-The software stack comes with the company’s Python-based TT-NN library, which can run AI models out of the box without needing to deal with the complexities of the hardware, so there is no need to write custom kernels unless the application requires it. 
+TT-Metalium’s bare-metal programming stack defines a low-level programming model to allow users to write custom kernels for deployment on Tenstorrent hardware.
+The software stack comes with debug and performance profiling tools as well as Tenstorrent's Python-based TT-NN library, which can run AI models out of the box without needing to deal with the complexities of the hardware. This negates theneed to write custom kernels unless the application requires it.
 
 TT-Metalium will be used to fine-tune performance for production deployments, where small efficiencies can be multiplied by the scale of the deployment, and to develop custom kernels for scientific computing, where operations are not well-supported by existing kernel libraries.