dScribe AI Sends Drones to Count Your Dirt Piles and That Is a Real Business

The Macro: Bulk Inventory Is a Guessing Game Worth Billions

There is a category of business problem that sounds boring until you learn how much money is at stake. Bulk inventory measurement is that category.

If you run a feed distribution business, you have piles of grain sitting in outdoor lots and indoor storage facilities. You need to know how much is in each pile so you can plan deliveries, reconcile accounts, and manage cash flow. If you run a quarry, you have piles of crushed stone, sand, and gravel. Same problem. Mining operations, construction material yards, recycling centers. Any business that deals in bulk materials that get stored in irregular piles faces this issue every day.

The current methods are genuinely bad. The cheap option is to eyeball it. A site manager walks around the pile and estimates the volume based on experience. This is how most small operations do it and the error rate is exactly what you would expect from asking a human to estimate the volume of an irregularly shaped pile of rock. Twenty to thirty percent off is normal. The expensive option is to hire a surveyor. A professional comes out with GPS equipment, takes measurements, does the math, and delivers a report. This is more accurate but costs thousands of dollars per survey and takes days or weeks. Neither option gives you real-time visibility into how your inventory changes between measurements.

The financial impact is not abstract. If your grain pile is actually 25% larger than you thought, you have capital sitting idle that is not showing up on your balance sheet. If it is 25% smaller, you are going to miss a delivery commitment and damage a customer relationship. Mining companies have reported inventory discrepancies worth millions of dollars because of measurement inaccuracy. Insurance claims, tax filings, and logistics planning all depend on inventory numbers that are, for most companies, educated guesses.

Drone-based surveying is not new as a concept. Companies like Propeller Aero and Kespry (now part of Near Earth Autonomy) have offered drone surveying services for years. But most of these solutions require trained pilots, manual flight planning, and post-processing that takes days. The technology has been available. The workflow has been the bottleneck.

The Micro: Feed Industry Experience Meets Computer Vision

Warren Wijaya Wang and Cole Robertson founded dScribe AI out of Olathe, Kansas. Warren’s background is relevant in a specific way. He worked at Milk Moovement, where he spent time with feed suppliers and farm managers figuring out logistics planning for dairy operations. He saw the inventory measurement problem from inside the industry, not from a drone lab. Cole was Head of Engineering at an AI startup and was engineer number three at Tropic, where he helped scale the company from seed through Series B. Technical depth on one side, domain knowledge on the other.

They came through Y Combinator’s Summer 2025 batch and raised $1.2 million from Abstraction Capital, Flyover Capital, Redbud VC, and KCRise Fund. The investor profile is interesting. Flyover and Redbud are Midwest-focused funds, which makes sense for a company selling to agricultural distributors and mining operations in flyover country. This is not a San Francisco pitch. The customers are in Kansas, Iowa, and West Virginia.

The product works in two modes. For outdoor stockpiles, autonomous drones fly predetermined routes, capture imagery, and build 3D reconstructions using computer vision. The system calculates precise volumes and weights from the 3D model. For indoor storage, fixed cameras do the same job without requiring a drone flight. The output is a dashboard showing real-time inventory levels, delivery and consumption tracking, and usage forecasting for logistics planning.

They have already landed a major agricultural distributor as a customer, which is a strong signal. Agricultural distributors are not early adopters by nature. They buy from companies that solve immediate, painful problems. The fact that dScribe closed a deal in this market suggests the pain point is as acute as the data says it is.

The positioning as a volume and weight measurement platform rather than a “drone company” is smart. Nobody in agriculture cares about drones. They care about knowing how much feed they have. The drone is an implementation detail.

The Verdict

I think dScribe is building in exactly the right way. They found a painful, expensive problem in a large industry. They have domain expertise and a technical cofounder who has scaled a startup before. They raised from investors who understand their customer base. And they are already generating revenue from enterprise customers.

The competitive question is whether the drone surveying incumbents adapt fast enough. Propeller Aero has distribution in mining and construction. If they add real-time monitoring and better automation, dScribe will be competing against an established brand. The advantage dScribe has is focus. Propeller is a general drone analytics platform. dScribe is specifically building for bulk inventory with features like usage forecasting and delivery tracking that only matter if you are managing commodity stockpiles.

In 30 days I want to see accuracy numbers. How does drone-based 3D volume measurement compare to professional surveyor results on the same pile? That is the data point that closes deals with CFOs. Sixty days, the question is how repeatable the autonomous flight workflow is across different site configurations. A flat lot in Kansas is different from a mountain quarry in Colorado. Ninety days, I want to understand the pricing model and whether it scales for smaller operations. A large distributor with twenty storage sites can afford a premium. A single-site feed lot in rural Nebraska needs a price point that competes with the cost of a guy walking around and guessing.