Building high-performance applications for AI hardware involves many moving parts: model export and compilation, pipeline configuration, performance tuning, analytics and output visualization. Voyager Wingman is designed to collapse that learning curve by giving developers an expert assistant with pro-level access to Axelera's SDK, documentation, and full software repo.
Voyager Wingman focuses on four areas where developers spend the most time:
End-to-end application building. Developers can describe the application they want — a model, its pre-processing and post-processing steps, or a cascaded multi-stage pipeline — and Wingman helps assemble a working computer vision pipeline, while maintaining the high accuracy expected.
Performance optimization. Wingman suggests concrete ways to speed up applications, recommends compiler configurations, and helps developers get the most out of Axelera hardware.
Debugging. Wingman helps diagnose common issues, such as configuration errors, device enumeration problems, and models that fail to compile, and walks developers toward a working result.
Documentation on demand. Instead of searching through reference material, developers can ask targeted questions about supported operators, APIs, runtime behavior, and configuration syntax, and get clear, contextual answers with examples.
Voyager Wingman is available as a web-based chat service and as a standalone application, with a plugin or skill for your preferred AI coding framework coming soon. In each of these modes, its knowledge is designed to stay up to date. As Axelera ships new Toolkit releases, examples, and documentation, the assistant is built to reflect those changes, with no downloads or manual updates required. The web-based option also means developers can reach Wingman from any platform through a single link, lowering the barrier to getting started. As a skill or plugin, which is coming soon, Voyager Wingman will fit directly into your existing workflow with no impact on your developer's setup and without exposing sensitive code. Wingman helps developers get started faster and accompanies them throughout the development, from setup to production.
"Voyager Wingman reflects how we want developers to experience our platform with fewer roadblocks, faster answers, and more time spent building. Our goal is to make world-class edge AI development accessible to every developer, not just experts." said Bram Verhoef, VP of Customer Engineering and Success.
Voyager Wingman is available through the Axelera Developer Community and Customer Portal. The chat is available under a freemium offering, providing developers with a credit allowance to explore the assistant at no cost. The app is free using bring-your-own-key (BYOK).
Voyager Wingman is the latest step in Axelera AI's mission to make high-performance, energy-efficient AI inference at the edge simple to adopt and deploy.
Axelera AI is a European AI semiconductor company developing next-generation AI acceleration hardware for edge computing and data center applications. With an edge-first architecture that addresses AI’s critical energy and cooling constraints, Axelera AI has deployed across more than 500 customers including telecommunications, aerospace, and enterprise sectors. The company’s Europa and Metis platforms deliver high-performance AI inference solutions that enable deployment at scale within real-world power and thermal budgets. Through its Partner Accelerator Network and Voyager SDK, Axelera AI provides a comprehensive ecosystem approach that ensures accessibility and rapid deployment. Axelera AI is supported through equity, grants and other non-dilutive funding instruments by leading funds including Innovation Industries, BlackRock, Samsung Catalyst Fund and other top-tier institutional and strategic investors.
Matt Stubbs
VoxusPR