Ananya Sharma

Pivoting from pure academia to building an AI startup in Bengaluru was both terrifying and thrilling. A year ago, running high-parameter LLMs efficiently on edge devices felt like a distant goal. After six months of failing to optimize memory bandwidth, our breakthrough came by rethinking tensor parallelism at the hardware interface level. Building tech in India's silicon valley means constantly pushing boundaries with world-class peers. The journey isn't just about training bigger models; it's about solving real latency issues for millions of local users. #ArtificialIntelligence #DeepTech #S…

View on Bcommune