Agrograde builds machines that sort onions and potatoes by quality, not just size
The Pune-based company makes AI-powered optical grading equipment for packhouses, designed around Indian crop varieties that imported sorters have struggled with.
Founded in 2017, Agrograde, formerly known as Occipital Technologies Private Limited, is a company based in Pune that specializes in post-harvest technology and agricultural robotics. The company's mission is to settle disputes over quality in the agricultural supply chain by building machines that sort onions and potatoes not just by size, but by quality.
Agrograde's CEO, Kshitij Thakur, and co-founder, Rakeshkumar Barai, have noted that over 99% of Indian packhouses still rely on manual quality checks, grading, sorting, weighing, and packing, with the workforce averaging an age of 45. This reliance on manual labor leads to inconsistencies in identifying and grading produce, resulting in rejected consignments and debit notes being sent back through the supply chain.
The company's flagship product, WaveMotion, is specifically designed for sorting onions and operates using a wave motion to accommodate the delicate outer skin of the onions, preventing damage. Vector Series, on the other hand, is Agrograde's inspection layer, which analyzes texture, boundaries, and context to distinguish defects from natural variations.
This approach allows the machines to sort unwashed, field-fresh onions and potatoes with up to 96% defect detection accuracy, without the need for recalibration when switching between potato varieties.
Thakur emphasizes that the company's systems are tailored to the unique physical behavior of each crop, rather than relying on camera-based sorting methods that have failed in Indian packhouses due to their inability to handle the high level of variation and the diverse crop varieties present in the country. In doing so, Agrograde aims to usher in the era of the autonomous packhouse, with grading being the first stage of the process.
Written by urgent.news from YourStory's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.