Real-World Supply Chain AI Applications with Gather AI's Sankalp Arora
🎯 Summary
Podcast Episode Summary: Real-World Supply Chain AI Applications with Gather AI’s Sankalp Arora
This 52-minute episode of Logistics of Logistics features an in-depth discussion with Sankalp Arora, Co-founder and CEO of Gather AI, focusing on how their technology bridges the critical gap between the physical reality of warehouse operations and the data recorded in Warehouse Management Systems (WMS). The core narrative revolves around how traditional, manual tracking methods lead to “leaky processes,” resulting in costly errors, delayed shipments, and operational inefficiency. Gather AI addresses this by using computer vision powered by AI to digitize workflows in real-time.
1. Focus Area
The primary focus is on real-world AI and Computer Vision applications within warehouse and supply chain operations, specifically addressing inventory accuracy, process digitization, and improving On-Time In-Full (OTIF) performance. The discussion centers on moving warehouse management from a reactive, inspection-based model to a proactive, process-monitoring model.
2. Key Technical Insights
- Real-Time Workflow Digitization: Gather AI uses commercial off-the-shelf cameras mounted on forklifts, drones, and docks to capture every movement (putaway, picking, staging, inbound/outbound). This captures what actually happened—including item location, quantity, and condition—rather than relying solely on manual scans.
- AI-Powered Error Prevention and Correction: The system detects errors (e.g., placing an item in the wrong bin) as they occur. It alerts the operator via a screen (e.g., on a forklift) for immediate correction, or it updates the WMS record to reflect the actual location, ensuring subsequent picks are successful even if the initial transaction was erroneous.
- Drone Efficiency for Inventory: Utilizing American-made drones, the company achieves 25 times faster inventory counting compared to manual methods, significantly reducing the need for costly, slow, and error-prone human cycle counting, often performed after hours.
3. Business/Investment Angle
- Inventory as a Major Liability: The discussion highlights that poor warehouse operations can result in retailers carrying 5% to 15% of their annual revenue in excess inventory, emphasizing the massive financial impact of inaccuracy.
- OTIF Pressure: Customer demands necessitate near-perfect OTIF rates (approaching 99.99%), forcing warehouses to evolve from being simple cost centers to critical performance drivers.
- Labor Attraction and Retention: Implementing advanced AI technology makes warehouse jobs more appealing to the modern workforce, positioning the role as technologically advanced rather than purely physical labor, which helps combat high churn rates (cited as up to 150% annually in some facilities).
4. Notable Companies/People
- Sankalp Arora (Gather AI): Co-founder/CEO, background in Robotics (Carnegie Mellon Ph.D.), involved in building the world’s first fully autonomous helicopter. His motivation for founding Gather AI stemmed from realizing robots are exceptional physical data-gathering machines applicable to civil society problems like supply chain inefficiency.
- NFI: Mentioned as a case study where Gather AI drastically improved operations for a high-velocity customer (inventory turning over four times in six months). Gather AI reduced inventory counting hours by 75% and reduced the time to locate lost pallets from days to minutes.
5. Future Implications
The industry is moving away from post-facto inspection and auditing (like traditional inventory control/QA) toward proactive process monitoring and control. AI and physical intelligence are enabling this shift, allowing companies to embed quality directly into the workflow rather than trying to inspect quality in later. This transformation is expected to become the standard for modern, high-speed fulfillment operations.
6. Target Audience
This podcast is highly valuable for Supply Chain Executives, Warehouse Operations Managers, Logistics Directors, and Technology Leaders interested in practical, deployed AI solutions that deliver measurable ROI in physical operations. It is also relevant for Venture Capitalists tracking innovation in industrial automation and logistics tech.
🏢 Companies Mentioned
💬 Key Insights
"So suddenly the digitization of your workflows will lead to automatic optimization using generative AI techniques."
"As we digitize each of your workflows in the warehouse, I think the as we become the source of truth of what's happening in your warehouse and your operations, you will see a team of AI agents from us working on that data and making decisions on your behalf or making your recommendations on how to improve your warehouse on a day-to-day basis."
"For them, the problem we are solving is bridging the gap between what their WMS says and what actually happens in the warehouse so that they can increase the throughput of their warehouses, increase on-time in-full while reducing the pressure on their labor."
"Usually that feedback loop has taken two or three months, even with digital twins. Our feedback loop reduces it to you can make intraday decisions on things that are happening or let our AI make those intraday decisions for you."
"For the first time, we are in partnership with Bosma and Austin Lighthouse. We are giving our the opportunity for visually impaired people to hold quality assurance jobs and inventory control jobs."
"I think being able to help people getting out of those environments and instant sending technology out is exactly how technology should play a role in our society."