We’ve published the second episode of Conversations with the CTO, featuring Simplex Micro CTO Thang Tran discussing a major engineering milestone: Dhrystone, Matrix Multiply, TinyML, and Qwen have all run successfully to completion on the Simplex performance model.

The conversation explores why executing a benchmark is only the first step, how RTL correlation validates true hardware performance, and why Qwen serves as a critical test for modern AI architectures. Thang also breaks down the role of memory latency tolerance and details how customers can utilize our C++ performance model to evaluate their own system-level workloads during hardware-software co-design.

The video is now available on the Simplex Micro YouTube channel.

Watch the video: https://www.youtube.com/watch?v=Za1FpAK8Uyg