Refining & Freeness
Refining and freeness control.
Pulp & Paper
Pulpum
Autonomous operations layer for pulp and paper manufacturing — stock prep, wet-end chemistry, forming, pressing, drying, coating and reeling, with web-defect vision and break prediction.
Overview
What Pulpum does.
Pulpum closes the loop on the paper machine: refining and freeness control, wet-end chemistry, forming, pressing, drying, coating and reeling, with web-defect vision, break prediction and an as-run machine simulation.
Every reel produces labelled process and defect traces that train the next model, tightening the loop from stock to finished reel.
| Detail | Value |
|---|---|
| Stage | Seed |
| Category | Physical AI |
| Year | 2026 |
| Website | pulpum.com |
| Headquarters | London, United Kingdom |
| Raised | $1M |
| Investor | 10xC |
| Model | Per-machine software subscription with an uptime and yield outcome component. |
| Measured by | Machine uptime, break rate, defect escape rate, reels expanded per customer. |
Products
One loop, module by module.
Every deployment on pulpum.com writes labelled traces back into the model, so the system generalises across plants, alloys, grades and geometries.
Chemistry & Retention
Wet-end chemistry and retention control.
Forming & Pressing
Headbox, slice and press control.
Drying & Coating
Steam, hood, coating and calender control.
Web Vision & Breaks
Web-defect vision and break prediction.
CD Profile Control
CD basis-weight, moisture and caliper control.
Broke & Grade
Broke, energy and grade sequencing.
Reel & Winder
Reel, winder and roll logistics.
Genealogy & Release
Per-reel genealogy and release.
As-Run Simulation
As-run machine simulation.
Why we invested
The thesis behind Pulpum.
A paper machine runs a mile a minute and a web break costs hours; defects are found after the reel is wound, and wet-end chemistry is tuned by the few people who know the grade. Paper volume is consolidating into fewer, larger machines.
Pulpum moves web vision and break prediction onto the line. The buyer is a paper mill paying in hard currency; the wedge is one wet-end or inspection step, and machine uptime and defect escape are the numbers that open the next module.
Who is building it
Sameer Khan
Co-Founder & CEOSameer owns the mill relationships, the pilot programme and the commercial model.
Asiya Bibi
Co-Founder & CTOAsiya owns the web-vision and process models, the edge runtime and the write-back into machine control.
Building something similar?
Building in Physical AI? We want to hear about it.