Tesla-style OTA · Local AI · Cloud control plane

Smart conveyor belts,
taught with samples.

Drop objects on the belt. Show SorterAI a few sample photos. It learns your sort rules — colour, shape, custom labels — and routes them to the right bin in milliseconds. Linux on the edge, Next.js in the cloud, secure mTLS in between.

≥ 98%
Sort accuracy
120 items / min
Throughput
< 90s
Update rollback
CAM
LeftMiddleRightReject
sorted: 0
live

A factory floor in your browser

Manage one belt or a hundred from the same dashboard. Push new rules in seconds. Roll back broken updates automatically.

Hybrid AI inference

Run YOLO + CLIP locally on the Jetson, or fall back to Anthropic / OpenAI vision models routed through our secure proxy.

Sample-driven learning

Upload a handful of reference photos. SorterAI builds CLIP embeddings and matches live frames in under 50 ms.

Visual rules engine

Compose conditions like Lego blocks: ‘red AND round → reject’. Edit rules from the touch panel or your laptop. Hot-reloaded in 2s.

Tesla-style OTA

Signed system bundles, A/B partitions, instant rollback if a health check fails. Canary, staged or fleet-wide.

mTLS device identity

Each belt provisions an ED25519 keypair on first boot. 90-day cert rotation, full audit log, JWT-signed commands.

Open hardware

Raspberry Pi 5 + optional Jetson Orin Nano + ESP32 bridge for hard real-time. Bill of materials in the docs.

From bricks to bins in four steps

  1. STEP 01

    Pair the belt

    Power on, scan the QR, enter the 6-char code from the touch panel.

  2. STEP 02

    Upload samples

    Drop a few photos of each object class. We embed them with CLIP.

  3. STEP 03

    Author rules

    Visual editor: ‘color is red’, ‘shape is round’, route → lane.

  4. STEP 04

    Push & monitor

    Live cam, throughput stats, alerts. Rollback any change in one click.

Two builds. One platform.

Start lean with cloud-AI. Upgrade to local inference when you need offline reliability or sub-50 ms decisions.

Pi 5 + ESP32 bridge
Soft-realtime in Rust + hard-realtime via embassy-rs
Optional Jetson Orin Nano
On-device YOLOv8n / CLIP, ~30 fps
12 MP autofocus camera
LED ring + IR trigger barrier for crisp captures
3 servo flaps
DS3218MG 20kg, ~150ms full-throw deflection
Pilz E-stop
Hardware-cut on the 24V rail. No software in the path.
Build · Pro
~1.400 €
Pi 5 + Jetson Orin + full BOM
Build · Basic
~700 €
Pi 5 only · cloud AI

Bring it online today.

Free tier includes 1 belt and 200 inferences / day. Upgrade only when you scale.