below marketmost similar roles pay hereabove market
This role pays more than
89%
of similar roles. Most pay
$177,250–$244,125
— the blue band above.
At the midpoint, this role pays about
$255k
versus about
$211k
for comparable roles.
Based on 240 similar postings.
Employer
About Amd
AMD (Advanced Micro Devices) is a semiconductor company that develops high-performance processors, graphics cards, and adaptive computing solutions for gaming, data centers, and embedded markets. Industry: Semiconductors
Amd currently has
496 open roles
on FindRole.
Listed pay typically runs
$172,000–$258,000
across 496 roles with salary data.
As an AI Research Scientist, Hardware AI Systems, you will join the team to develop AI systems that generalize across various hardware engineering contexts, including different SoC programs, toolchains, and process nodes. Your primary responsibilities involve researching transfer and multi-task learning for engineering agents, developing metareasoning policies for budgeting expensive EDA steps, and building cross-program benchmarks to track generalization gaps. You will work on representation learning across RTL, verification, and physical design flows while addressing high-latency rewards typical of silicon signoff. The role requires deep technical expertise in machine learning, specifically transfer learning, meta-learning, intelligent agents, and long-horizon reinforcement learning. You will also utilize Large Language Models for tool use and planning to solve complex problems within hardware development and silicon design workflows.
Research transfer and multi-task learning for engineering agents across various IP blocks and tools.
Develop metareasoning policies for budgeting EDA steps and selecting appropriate abstractions.
Build cross-program benchmarks and datasets to identify and track generalization gaps.
Collaborate on reward shaping and credit assignment for slow or heterogeneous feedback systems.
Turn one-off wins into reusable models, benchmarks, and transfer protocols for the hardware roadmap.
Publish research at top venues and maintain internal technical standards for evidence-backed generalization claims.
What we're looking for
Research transfer and multi-task learning for engineering agents across IP blocks, tools, and program generations.
Develop metareasoning policies for budgeting EDA steps, selecting abstractions, and recovering from tool or data failures.
Build cross-program benchmarks and datasets to expose generalization gaps and track progress over time.
Collaborate with RL scientists on reward shaping and credit assignment for slow or heterogeneous feedback.
Publish at top venues and maintain internal technical standards for evidence-backed generalization claims.
Deep technical expertise and proven research experience in ML, emphasizing transfer learning, meta-learning, intelligent agents, and long-horizon RL (preferred).
Exposure to hardware development (RTL, verification, PD, or bring-up) or willingness to ramp with embedded experts (preferred).
Experience with LLM tool use, planning, or hierarchical control in real toolchains (preferred).
PhD in Computer Science, Machine Learning, or related field (preferred).
Embedded Systems
Robotics
Mechatronics
Firmware
CAN
Ethernet
FMEA
V&V
Root-cause Debugging
Sensor Suites
Power Distribution
Motion Control
ICDs
Systems Engineering