NAVER D2SF invests in AI startup TraceWeave for on-device applications
NAVER D2SF has completed an investment in TraceWeave, a developer of a post-Transformer AI architecture designed to operate on memory-limited hardware.
Investment firm NAVER D2SF has backed TraceWeave, an early-stage startup focused on optimizing AI processing for edge devices. TraceWeave has engineered an alternative to standard Transformer models, aiming to enable AI agent functionality on hardware with constrained memory capacity.
The startup recently concluded a paid pilot project within the mobility sector to test the viability of its architecture. This development is part of a broader industry shift toward running artificial intelligence locally on hardware—such as drones, robotics, and smart eyewear—rather than relying exclusively on cloud-based computing.
NAVER D2SF continues to prioritize the on-device AI segment, viewing the technology as a bridge to expanding AI utility in offline environments. TraceWeave now intends to transition its research into commercial applications across various hardware platforms.