TeknTrash effectively *recycles data*: it identifies products at recycling centers and matches their origin to the stores where they were sold.
Running on own funds
According to the World Bank, the Household final consumption expenditure (HFCE) of is US$13 billion in USA and US$9 billion in Europe: 68% and 56% of GDP accordingly. These markets are made of the companies we plan to tackle.
Problem or Opportunity
Historically, product manufacturers have known the history of their products only until the store. After that, nothing was known on where, what, how, etc they were sold and fully consumed
The end result is unnecessary stock, incorrect size packaging, use of plastic, not knowing anything about the competitor, etc
Solution (product or service)
Tekntrash “recycles” data. That is, it obtains data fr om residues
Its military grade ML optical recognition is able to identify products at disposal sites and match its origin all the way to the store wh ere it was sold
That allows for a better understanding on consumer habits and thus more sales and less costs.
Advantages or differentiators
Our technology is based on a 3 step recognition system
1 – A contract is closed with a company which wants to know how its product is being used
2 – The product is photographed under any circumstance and with the closest environment or background possible, in order to avoid inference and labeled using systems such as Figure Eight
3 – The product is trained using GPUs (Nvidia) or TPUs (Google cloud) to speed up the process
4 – The trained system is fed into the server identifying products at disposal sites or the reverse vending machines. This server uses TeknTrash identification technology to identify it when it passes and count its occurrence
5 – The server data is matched to the recycling center data and the customer sales data to create a full lifecycle analysis of the product