Prithviraj Banerjee on numerical methods
From S2, EP6 - Dr. Prith Banerjee - ANSYS CTO · · DrNeilAshton
“In our world the problem is the tradeoff between accuracy and speed; we want both accuracy and speed and for the tool to be easy to use and converge all the time — that's the holy grail. My job as CTO is to look at solvers and ask how can I make them more accurate, faster, easy to use and converge reliably.”
On , Prithviraj Banerjee, Chief Technology Officer at ANSYS Inc, spoke about numerical methods during S2, EP6 - Dr. Prith Banerjee - ANSYS CTO on DrNeilAshton.
About Prithviraj Banerjee
Prithviraj Banerjee, Chief Technology Officer at Ansys, discussed the company's approach to simulation and artificial intelligence in a December 2024 podcast. He described Ansys as a leading modeling and simulation company that uses physics-based methods such as finite element and finite volume techniques. Banerjee stated that his role involves examining the future of simulation and technologies like AI, machine learning, and high-performance computing to drive future products. He noted that a key challenge is balancing accuracy and speed in simulation tools. Banerjee outlined Ansys's development of a platform called SimAI, which allows customers to train AI models on their own design problems, reducing simulation time for later designs. He also described a longer-term vision for foundational AI models for physics, analogous to large language models, that would be trained by Ansys on a broad set of physical data. Banerjee acknowledged that building such models would require access to customer data and discussed potential business models involving anonymized data sharing with compensation. He also mentioned that quantum computing could accelerate simulation within a decadeasi.