What do you still count manually at work?

I’m building SmartCheck — a browser-based tool that counts similar objects in a photo after you tap just 3 examples.

No model training. No app install. Just open it in your browser, tap 3 objects, and let AI do the counting.

What surprised me while building it:

Getting the number isn’t the hardest part.
Trusting the number is.

In real-world work, you need to know:

→ What did the AI actually count?
→ Did it miss anything?
→ Did it count something it shouldn’t?

That’s why SmartCheck shows every detected position directly on the image — and lets you add or remove detections before using the final count.

I’d love to learn how people are handling this today 👇

1️⃣ What do you still count manually at work?
Inventory? Materials? Products? Vehicles? Crops? Something completely different?

2️⃣ What makes it frustrating?
Too slow? Easy to miscount? Hard to double-check?

3️⃣ What would make you trust AI-assisted counting?
Visible detection points? Manual correction? Accuracy reports? Something else?

📷 If you have a non-sensitive example photo or use case, feel free to share it below.

I’ll try some of them with SmartCheck and share the results here — including the cases where it doesn’t work well.