AMD launches Ross AI assistant for embedded developers

AMD has launched Ross, an AI assistant for embedded developers. The product is available now.

Ross is designed to support the full embedded development process, including hardware and software design, optimisation, debugging, AI implementation, and deployment. Built for AMD's embedded development environments, it connects directly with the company's tools, documentation, and engineering workflows.

The launch reflects growing use of AI tools in chip and systems engineering, where development teams are managing increasingly complex combinations of silicon, software, board design, and machine learning. AMD's embedded portfolio covers FPGAs, adaptive SoCs, x86 embedded processors, and edge AI platforms, each with its own toolchains and design processes.

Ross uses natural-language interaction to let developers search documentation, check tool status, run commands, and execute predefined workflows. It also supports engineers' preferred large language models, integrated development environments, and command-line environments while remaining linked to AMD Embedded tools.

AMD said the assistant is built around four elements: model context protocol servers that connect AI agents to development tools, an AMD Knowledge Base drawn from company documentation, agent skills written by experts, and design examples showing how those workflows apply in embedded applications.

According to AMD, Ross can help users with hardware and software partitioning, high-level synthesis-based hardware design, silicon optimisation and debug, embedded software development, machine learning optimisation, power estimation, and board-level tasks such as schematic review and layout.

Salil Raje, Senior Vice President and General Manager, AMD Embedded, said the aim is to reduce friction across a broad and increasingly technical development cycle.

"Embedded development is becoming increasingly complex as teams work across hardware design and debug, software development and deployment, AI inference, and system-level design. 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," said Salil Raje, Senior Vice President and General Manager, AMD Embedded.

AMD is positioning Ross as a specialist product rather than a general-purpose coding assistant. The system combines access to development environments with validated documentation and reusable engineering knowledge, allowing teams to automate repeatable tasks and apply established internal methods more consistently.

Workflow focus

Examples provided by AMD include using natural-language prompts to search technical material, execute tool commands, generate code, and revise designs. The company added that Ross can explain tool errors, analyse timing issues, suggest implementation changes, improve Vitis HLS designs, and support hardware debugging through signal capture and interface analysis.

This focus on workflow automation may appeal to engineering groups looking to shorten design iteration cycles and reduce onboarding time for less experienced staff. AMD said the product is intended to make specialist knowledge easier to reuse across teams and projects, while allowing senior engineering methods to be codified into repeatable agent skills.

AMD also included a customer comment from iWave Global, which has used the assistant in development work.

"The AMD Ross agentic AI assistant has been a valuable addition to our development workflow. It helps us identify and troubleshoot issues more efficiently, reducing the time and effort required during the debugging and bring-up stages," said Geetha Govindaraj, Associate Director, FPGA SOM BU, iWave Global.

AMD said Ross will expand over time to cover additional embedded tools and workflows. At launch, the product forms part of a broader push by semiconductor companies to embed AI more deeply in specialist engineering software, particularly in areas where design complexity, documentation volume, and debugging demands can slow development work.

According to AMD, Ross operates within configured user permissions and supported tool environments.