Shodh AI

Accelerating the Decade-Long Discovery Cycle

Topics

  • The development of a new material generally requires a decade and incurs costs amounting to millions of dollars. Shodh AI aims to significantly reduce this timeline by an order of magnitude. Under the leadership of Dr Arastu Sharma, the company is developing Project Skanda, a 7-billion-parameter model designed for materials discovery that incorporates artificial intelligence into experimental processes. 

    This alpha release, unveiled at the India AI Impact Summit, targets the mesoscale, which is the structural architecture inside a battery cell that determines its lifespan, safety, and charging speed. This is a meaningfully different approach from models focused on atomic chemistry, and positions Skanda in territory that few global competitors currently occupy. 

    The platform runs on NVIDIA’s Hopper GPU infrastructure and the PhysicsNeMo framework, and is supported by the IndiaAI Mission’s AI for Science mandate. For investors tracking climate technology, battery innovation, and deep-tech AI simultaneously, Shodh AI sits at the intersection of all three.

    Topics

    More Like This

    You must to post a comment.

    First time here? : Comment on articles and get access to many more articles.