EEA shortlist: Analog Devices

Integrating AI/ML models into embedded applications is often a complex and fragmented process. Developers face challenges in optimising models for hardware constraints, managing middleware, and gaining real-time performance insights. These hurdles result in extended development cycles, suboptimal performance, and limited scalability. The prototype of CodeFusion Studio (CFS) AI Fusion Tools addresses these issues by streamlining AI model deployment within a multi-purpose embedded development platform, including tools enabling embedded developers to get more insight into how their model is running on ADI hardware. 

With real-time model analysis and integration with the ADI Model Zoo, AutoML for Embedded and the Data Extraction Framework software stack, CFS AI Fusion Tools are designed to drastically reduce development complexity and accelerate time-to-market for intelligent embedded solutions.

The CFS AI Fusion Tools prototype is uniquely innovative due to its holistic, hardware-aware approach to AI deployment. It streamlines the development and deployment of AI/ML models by combining advanced compilers, hardware-specific optimisers, and real-time inference analysis within a single, integrated platform. Unlike generic AI toolchains, CFS AI Fusion is purpose-built for embedded systems, offering deep integration with ADI hardware accelerators and Data Extraction Framework middleware. This synergy enables precise performance tuning and dynamic model validation, allowing developers to deploy AI solutions that are both efficient and scalable. CFS AI Fusion transforms the traditionally siloed AI development workflow into a streamlined, feedback-driven process that maximises embedded system performance.

For more information, visit: www.analog.com/codefusionstudio

The Electronics Excellence Awards ceremony will take place at embedded world on 12th March at 4.30pm at the Hall 3 Exhibitor’s Forum.

You’re all welcome to come and see which product will win!