Skip to main content
Skip to content
Log in

projects

+ New project
Edge AI on the Edge
deveco.io project

Edge AI on the Edge

Oumuamua Digital Marketing Technology Solutions is a Technology Company . That has developed a Verticalized, Sovereign Air-Gapped, Asymmetric Neuro-Symbolic Cognitive Edge Appliance and Autonomous Cyber-Physical Orchestration Ecosystem (The IC-COS Platform Stack).This technology establishes an entirely new, independent class of mobile machine intelligence. By permanently trapping advanced civilizational reasoning directly within local, disconnected silicon, our platform successfully achieves a multi-decade computer science milestone: the complete, unconditional decoupling of high-fidelity cognitive computation from cloud data centers, public network routing, and third-party infrastructure dependencies. The technology represents a complete structural departure from current infrastructure-dependent cloud and centralized computing models. By successfully automating the boundary between dynamic context-injection control layers and proprietary frozen neural weights, this system achieves a critical industry milestone: it decouples advanced, high-fidelity cognitive reasoning from the datacenter and traps it permanently within local, offline silicon.The platform requires no physical semiconductor fabrication, custom hardware manufacturing lines, or public data-center real estate allocations. It operates as a pure software-defined fabric designed to optimize and secure an acquiring organization’s existing physical computing infrastructure fleets. A technical administrator can initialize the complete air-gapped system, deploy our customized context frameworks, and have a live, token-streaming engineering engine running on current enterprise laptops and workstations in under 5 minutes.

PlantVillage Disease Classification
deveco.io project

PlantVillage Disease Classification

This project walks through the complete pipeline to train, quantize, convert, and benchmark an AkidaNet model on the PlantVillage dataset for Akida 1 hardware. PlantVillage contains 54,303 images of healthy and diseased plant leaves across 38 categories (14 crop species × multiple disease types plus healthy variants). The task is a 38-class image classification problem: given a 224×224 RGB image of a leaf, identify the crop species and disease (or healthy state).