AMD launches ‘Ross’ agentic AI assistant to speed up embedded design, development
Semiconductor major AMD on Monday launched “AMD Ross”, an agentic artificial intelligence (AI) assistant designed to accelerate the complete embedded development flow from architectural design to system deployment.
The platform spans hardware design, optimisation, debugging, printed circuit board (PCB) design, schematic review, embedded software, edge AI implementation, and actual system deployment, according to the American semiconductor multinational.
AMD Ross enables natural-language tool interaction and deploys expert-developed agent skills, reusable engineering workflows, and a grounded AMD Knowledge Base to guide the entire design process.
The assistant connects to development tools across the AMD Embedded portfolio, covering FPGAs, adaptive SoCs, x86 embedded processors, and specialised edge AI platforms.
“Embedded development is becoming increasingly complex as teams work across hardware design and debug, software development and deployment, AI inference, and system-level design,” said Salil Raje, senior vice president and general manager, AMD Embedded.
Raje added, “AMD Ross brings AMD Embedded tools, trusted knowledge and expert-authored workflows together in a single agentic AI experience grounded in the technologies and methodologies our customers use every day.”
“AMD Ross brings the power of agentic AI to embedded developers to move product innovations from design intent to deployment faster by accelerating the entire life cycle,” Raje noted.
The client-agnostic assistant allows engineering teams to integrate their preferred large language models (LLMs), integrated development environments (IDEs), and command-line environments while remaining connected to AMD tools.
The architecture comprises Model Context Protocol (MCP) servers, AMD-validated vectorised databases, open-standard agent skills in markdown format, and ready-to-run design examples.
Using open MCP servers, the assistant queries information, executes commands, analyses timing issues, performs high-level synthesis optimisations, and assists hardware-software partitioning.
The platform enables developers to optimise machine learning algorithms, estimate power requirements accurately, and perform system schematic reviews and optimal board layouts.
By capturing institutional engineering expertise into reusable agent skills, the tool automates repetitive tasks, accelerates new engineer onboarding, and ensures consistent execution of best practices.
AMD Ross is available immediately, with the company planning monthly rollouts of additional AMD Embedded tools and workflow capabilities under the platform.
