LATEST NEWS

AMD Releases Ross Agentic Assistant for Vivado and Vitis HLS Workflows

The available-now engineering assistant connects language models to AMD development tools, documentation and reusable FPGA design workflows.
AMD has released Ross, an agentic AI assistant designed to support embedded-system development through natural-language interaction with AMD tools and technical documentation.
Ross initially provides workflows for the Vivado Design Suite and Vitis High-Level Synthesis. It can search AMD documentation, inspect active tool sessions, execute commands and guide users through design, optimisation and debugging procedures.
The system connects AI clients to supported engineering tools through Model Context Protocol servers. Engineers can use their preferred compatible large language model, integrated development environment or command-line client instead of being restricted to a single AI provider or front end.
Ross combines four principal elements: MCP servers for tool access, an AMD technical knowledge base, expert-authored Agent Skills and tested design examples. Agent Skills encode repeatable engineering procedures so that teams can reuse established approaches to tasks such as timing closure, design analysis and HLS optimisation.
The knowledge base can be accessed online, while AMD also provides multi-gigabyte local documentation packages. Local access can be useful in development environments where proprietary source code or design information cannot be submitted to an external knowledge service.
AMD positions Ross as supporting work across hardware design, embedded software, edge-AI implementation, power estimation and system deployment. The assistant can also read utilisation and timing reports, help diagnose failed paths and suggest changes to a design.
In one AMD internal HLS example, Ross was used to optimise an image-sensor demosaicing pipeline while meeting the same 300 Mpixel/s throughput target as a manually written implementation. AMD reported that the resulting design used 35% fewer lookup-table resources and was completed in two hours rather than two days. These results apply to the tested workflow and are not a general benchmark for every design.
Ross is available now, and AMD plans to add further embedded-development tools and workflow capabilities on a monthly basis. AI-generated commands and design changes still require engineering review, simulation and hardware validation before they can be used in a production system.


Image credit: AMD

Galia

Recent Posts

How AI Chips Are Redefining Semiconductor Reliability Testing

Teradyne’s Titan HP combines realistic device workloads, per-device junction-temperature control and a traceable stress history,…

12 hours ago

Security by Design: Fraunhofer Brings RISC-V Root of Trust into the Silicon

Following functional validation of a 22 nm FD-SOI prototype, Fraunhofer is offering a RISC-V and…

12 hours ago

Beyond the Traditional MCU: Rethinking Embedded Architecture for Edge AI

Synaptics’ Astra SR200 family combines heterogeneous Arm cores, Torq neural acceleration and package-integrated DRAM to…

12 hours ago

Beyond Speed: How imec’s 100 GHz Photodetector Enables 400 Gbps Optical Links

Belgian research institute imec has demonstrated a Ge/Si avalanche photodiode combining 100 GHz bandwidth, 1.8…

12 hours ago

Melexis Advances Core Body Temperature Measurement For Smart Wearables

Tessenderlo-Ham, Belgium, 11 November 2026 – Melexis introduces the MLX90636, the world’s smallest contactless temperature…

4 days ago

Powell Electronics Europe expands across Europe and further afield, delivering interconnect solutions for mission-critical applications

Hi-rel connectors and components backed by in-depth product knowledge Powell Electronics Europe, the supplier of…

4 days ago