One of the structural problems in recyclable materials trade that we have not yet discussed in this series: listings.
Not the existence of listings - platforms has tried to create listings for recyclable materials before. The quality of listings.
When you ask someone to describe their OCC in a free-text field, you get "Old Corrugated Containers, good quality, 200 ton." When you ask them to describe their HDPE natural, you get "HDPE bottles, clean, natural color."
These descriptions are not useful to a buyer. A buyer needs to know the grade designation, the contamination tolerance, the moisture specification, the bale weight and density, the purity ratio. The information that actually makes a purchasing decision possible.
The problem is not that generators and MRFs do not know this information. They know it. The problem is that every listing platform has asked them to type it from scratch, in a format that varies by platform, without any intelligence layer helping them describe their material in the terms buyers actually need.
WHAT KIWI AI DOES
Kiwi AI is the grade intelligence layer built into Inacsys. When a user selects their material grade - OCC-11, HDPE natural bales, PET clear flake, Double Sorted OCC - Kiwi AI already knows what that grade is. It surfaces the specification, the standard end uses, the accepted contamination thresholds, the synonyms and acronyms that buyers in different markets use for the same material.
The user confirms what is specific to their lot - quantity, location, condition, asking price. The listing builds from that knowledge, not from a blank form.
The same intelligence layer applies to RFQs. A buyer looking for a specific grade can describe their requirement through Kiwi AI's grade selection - and the resulting RFQ is described in terms that generators can immediately understand and respond to.
WHAT THIS MEANS FOR TRADE QUALITY
A grade-accurate listing creates a different kind of trade conversation. The buyer's first question is not "what exactly is this?" - because the grade specification already answers it. The first question can be about price, logistics, or timing.
This is not a small thing. In a market where every trade starts from zero, reducing the friction at the listing and discovery step has a meaningful impact on how many conversations turn into actual trades.
Kiwi AI is deployed in Phase 1 of Inacsys - available from September 5 for both listing creation and RFQ curation.
15 days of founding access remaining. Platform goes live September 5, 2026 globally.
? www.inacsys.com