Agricultural AI: Two More Kilograms Per Ton
Hosted by Findability Sciences
Wednesday, October 14, 2026
10:00 AM–11:00 AM ET
About
In the 2025 Pantaleon harvest, validated recovery moved from 81 to 83 percent, roughly two additional kilograms of sugar per ton, with record-low lost time.
The interesting part is not the result but the sequence that produced it. Before any model was built, the mill agreed a reference for its own potential, measured the distance to it, and put its Finance team in charge of the number. This hour walks that sequence: what gets measured at the field-to-harvest and harvest-to-factory handoffs, why purity sets the recovery ceiling, and the five layers that have to hold underneath any of it, being Infrastructure, Collection, Unification, Processing, and Presentation of Data, a framework recognized by IDC, Gartner, and Google. Throughout, the system recommends and the operator acts.
For mill managers, agricultural and factory leadership, and agro-industrial technology teams.
A talk by Sagar Mahurkar, Vice President at Findability Sciences, with live Q&A. Sagar is a TEDx speaker and IEEE award-winning researcher who holds 5 US patents and has worked on more than 60 AI projects himself, drawing on Findability Sciences' 250+ deployments as a Fortune America's Most Innovative Company.
Speakers
- Sagar Mahurkar is Vice President at Findability Sciences, where he leads accounts, technology, and investor relationships across the United States, Japan, and India. He works with manufacturers, retailers, pharmaceutical companies, financial institutions, and law firms to move enterprise AI out of experimentation and into daily operations: forecasting what will happen, then acting on what to do next.
- He is an engineer before he is an executive. He holds five US patents, and his inventions have shipped at consumer scale rather than stopping at the lab. One calibrates the barometric sensor in a phone to estimate which floor a 911 caller is standing on, so that first responders reach the right apartment. Another uses Wi-Fi signal alone to detect motion and count people inside a home, with no camera involved. A third secured credential sharing across 16.8 million devices. A fourth delivered in-home content to mobile devices across 6.3 million set-top boxes. He also holds an IEEE Best Paper Award, from UEMCON at Columbia University, for work integrating object detection with augmented reality.
- That record began at Verizon, where he won the company's MVP award for individual performance. He built augmented-reality applications that walked self-install customers through router and set-top box setup using live object detection, launched the first MEC-based AR/VR concerts, and worked on the floor-detection technology behind E911 location accuracy. Across that period he shipped applications serving between 6.5 million and more than 100 million users, which is where he learned the difference between a model that works and a system that survives contact with users at scale.
- He moved into enterprise AI in 2022, joining Findability Sciences as Director of Technology and becoming Vice President in 2023. He now leads live global engagements for accounts including Daikin, The Home Depot, and Mitsubishi, designing solutions that combine predictive, interpretive, and generative AI, and has worked on more than 60 AI projects himself. During his tenure the company has been named a Fortune America's Most Innovative Company for 2023 and 2024, US Government Microenterprise of the Year, and an Inc. Magazine fastest-growing company.
- His argument to enterprise audiences is consistent and slightly unfashionable: a large language model is one component of a system, not the system. Most organizations that conclude AI does not work for their business have in fact bought a model and skipped the architecture underneath it. What produces intelligence at scale is a factory, built layer by layer, and Sagar's work is largely spent identifying which layer a stalled project actually died on.
- Alongside the commercial work he keeps a foot in research. He reviews for five IEEE international conferences, publishes with IET, and serves on the Executive Advisory Board for Data Science and AI at Worcester Polytechnic Institute. He manages research relationships with WPI, MIT, and the University of Tokyo, and two years of collaborative work with WPI graduate students and faculty has produced both academic awards and capability inside Findability Sciences products. His online courses have reached more than 6,000 students.
- He has spoken at NASA and at MIT's Startup Exchange, appeared on enterprise AI panels for NASSCOM, IDC, NRF, and CIO100, and delivered a TEDx talk, "Bringing Dreams to Life with Augmented Reality," at TEDxWPI.
- Sagar holds a BS in Electrical and Computer Engineering with Honors from Worcester Polytechnic Institute and an MBA from the University of Illinois at Urbana-Champaign. He lives in the Boston area.
Host
Findability Sciences
Location
Findability Sciences Office at Workbar Burlington 100 Summit Dr, Ste 917, Burlington, MA 01880 3rd floor in the building
100 Summit Dr, Ste 917, Burlington, MA 01880
Open in Google Maps ↗