Jul 2024 – Present · 1 yr 11 mos · On-site

• Enabling GenAI at the edge.

Oct 2022 – Jul 2024 · 1 yr 9 mos · On-site

• I have been entrusted with the leadership and management of a dynamic team comprising approximately 15 to 20 professionals. Our primary focus has been the delivery of high-quality releases, facilitating the compilation of cutting-edge models, including generative AI models such as Stable Diffusion, onto edge devices.
• I have played a pivotal role in providing valuable feedback to the hardware team regarding the development of the second generation of the processor. This input has been instrumental in addressing the shortcomings observed in the first generation. Furthermore, I have been responsible for overseeing the rearchitecture of the kernel library, leading to a remarkable performance scaling of up to 5x.
• In addition to these technical responsibilities, I have actively nurtured and empowered my team, enabling them to scale their capabilities and contribute significantly to the resolution of customer issues.

Jan 2021 – Nov 2022 · 1 yr 10 mos · On-site

• I significantly enhanced the scheduler runtime, taking charge of the development of streamlined cost estimation methodologies to facilitate improved schedule generation for the chip. I also spearheaded a comprehensive compiler redesign, equipping it with the capability to handle dynamic inputs, support 3D processing, and enable multiple complex neural network operations. Additionally, I led the design of our customer-centric runtime validation process, prioritizing the highest quality standards.

Dec 2018 – Jan 2021 · 2 yrs 1 mo · On-site

• I took the initiative to spearhead the development of a machine learning compiler kernel library tailored for the inaugural processor, which played a pivotal role in driving company revenues. Commencing from its inception, I methodically engineered and fine-tuned this library to attain the highest echelons of efficiency.