Meltra
Melt chemistry and furnace control.
Metal Casting & Foundry
Smelteon
Autonomous operations layer for metal casting and foundry manufacturing — melt chemistry and furnace control, robotic pouring, AI X-ray and CT porosity inspection.
Overview
What Smelteon does.
Smelteon closes the loop on metal casting and foundry manufacturing: melt chemistry and furnace control, sand molding and coremaking, robotic pouring and mold fill, solidification control, AI X-ray and CT porosity inspection, robotic fettling and the as-cast digital twin.
Every pour produces labelled process and inspection traces that train the next model, tightening the loop from charge to finished casting.
| Detail | Value |
|---|---|
| Stage | Seed |
| Category | Physical AI |
| Year | 2026 |
| Website | smelteon.com |
| Headquarters | London, United Kingdom |
| Raised | $2M |
| Investor | 10xC |
| Model | Per-line software subscription with a sound-casting yield outcome component. |
| Measured by | Sound-casting yield, melt chemistry conformance, porosity escape rate, lines expanded per customer. |
Products
One loop, module by module.
Every deployment on smelteon.com writes labelled traces back into the model, so the system generalises across plants, alloys, grades and geometries.
Coreon
Sand molding and coremaking control.
Pourbot
Robotic pouring and mold-fill control.
Poreon
AI X-ray and CT porosity inspection.
Castwin
As-cast digital twin.
Fettlix
Robotic fettling and finishing.
Why we invested
The thesis behind Smelteon.
Foundries run on heat and judgement: a melt that drifts off chemistry is caught after the pour, and porosity is found at X-ray after the casting is made. The people who can read a melt are scarce and retiring.
Smelteon encodes the melt and the inspection. The buyer is a foundry paying in hard currency; the wedge is one melt or inspection step, and sound-casting yield is the number that opens the next module.
Who is building it
Abdul Basit
Co-Founder & CEOAbdul owns the foundry relationships, the pilot programme and the commercial model.
Malik Tariq
Co-Founder & CTOMalik owns the melt and defect models, the edge runtime and the bounded write-back into furnace and pour control.
Building something similar?
Building in Physical AI? We want to hear about it.