Agrobot startup Agrograde develops AI machines for tomatoes and garlic

Kshitij Thakur, founder and CEO of Agrograde, stated that the Pune-based startup currently provides such solutions for onions, potatoes, apples, and Areca palm. Developments for tomatoes may appear around next year. In addition, the company plans to enter the spice segment, and it will take another 2–3 years to create and assess the viability of such technologies.

The company was founded in 2018. To date, it has installed over 130 machines across 14 states and four value chains. Key clients include farmer producer organizations, aggregators, and exporters, as manual operations for sorting, grading, and packaging are becoming inefficient. The use of Agrograde machines has reduced sorting and grading costs for onions to 0.08–0.10 rupees per kilogram with a size accuracy of about 99% and reduced handling losses.

The equipment uses wave motion technology and deep learning. Each machine captures at least 12 images of the produce, which are analyzed by algorithms trained on data from the past 8 years. The mechanisms detect rot and fungal damage at a speed of about 10 tonnes per hour, completing tasks in a couple of hours instead of 1–2 days. The cost of the equipment ranges from 2.5 lakhs to nearly 1 crore rupees depending on the configuration, capacity, and custom settings.

Currently, manufacturing acts as a bottleneck due to the limited number of units produced per month. At the same time, the firm receives inquiries from abroad and intends to enter the Southeast Asian markets within 18 months after scaling. Agrograde also continues to deepen its work with the onion and potato supply chains, actively introducing equipment in the states of Maharashtra, Karnataka, Gujarat, Chhattisgarh, Kerala, and Tamil Nadu.