ML Systems Research Engineer, RL / Inference / Agent Systems Job Listing at AMD in Santa Clara, CA (Job ID 87871-en-us)

Description

WHAT YOU DO AT AMD CHANGES EVERYTHING 

At AMD, our mission is to build great products that accelerate next-generation computing experiences—from AI and data centers, to PCs, gaming and embedded systems. Grounded in a culture of innovation and collaboration, we believe real progress comes from bold ideas, human ingenuity and a shared passion to create something extraordinary. When you join AMD, you'll discover the real differentiator is our culture. We push the limits of innovation to solve the world's most important challenges—striving for execution excellence, while being direct, humble, collaborative, and inclusive of diverse perspectives. Join us as we shape the future of AI and beyond.  Together, we advance your career.  

THE ROLE
We are hiring ML Systems Research Engineers to build the distributed training, reinforcement learning, inference, model optimization, and agent infrastructure behind AI-for-engineering systems.

Candidates should bring strong depth in at least one of the following areas:

• Distributed training, reinforcement learning, or post-training systems
• LLM inference, model optimization, or GPU performance

Experience across multiple areas is valuable, but expertise in all areas is not required.
You will work across compute optimization, hardware engineering automation, verification, simulation, debugging. The emphasis is on scalable ML systems that make research practical, repeatable, and useful for production engineering teams.

THE PERSON
You are a systems-minded ML engineer or researcher who understands that model quality depends on the surrounding loop: data, tools, inference, graders, reward design, logging, and iteration speed. You can build reliable infrastructure, reason about RL and inference tradeoffs, and collaborate with scientists and applied engineers to make experiments reproducible and useful.

KEY RESPONSIBILITIES
Depending on your background, you may:

• Build distributed training, RL, post-training, inference, or model optimization systems.
• Improve training and inference efficiency through parallelism, scheduling, communication optimization, caching, checkpointing, and fault tolerance.
• Develop rollout, sampling, scoring, retry, experiment-tracking, and reproducibility infrastructure.
• Build systems for long-horizon engineering tasks where validation may take minutes or hours.
• Design reward pipelines, staged rewards, proxy graders, and methods for detecting reward hacking.
• Develop high-throughput, low-latency, and reliable LLM serving systems.
• Improve model performance through quantization, mixed-precision inference, kernel optimization, and performance tuning.
• Build model and agent pipelines that interact with compilers, profilers, simulators, validation systems, benchmark harnesses, and internal services.
• Develop correctness, performance, and regression evaluation for training, inference, and agent workflows.
• Analyze experimental results and translate system behavior into improvements for models, rewards, tools, and infrastructure.

TECHNICAL FOCUS AREAS
Distributed Training, RL, and Post-Training

• Distributed training and large-scale post-training
• Data, tensor, pipeline, sequence, or expert parallelism
• Checkpointing, fault tolerance, elastic training, and communication optimization
• RLHF, GRPO, PPO, DPO, preference optimization, and reward modeling
• Rollout generation, actor-learner systems, and trajectory management
• Sparse, delayed, mixed, or expensive rewards
• Training and inference colocation

Inference and Model Optimization

• LLM serving, batching, scheduling, sampling, and request routing
• KV-cache management and distributed inference
• Quantization, low-precision inference, calibration, and accuracy-performance tradeoffs
• Kernel, runtime, memory, and GPU performance optimization
• Tool execution, sandboxing, retries, and recovery
• Long-horizon workflow orchestration and state management
• Reliability, observability, capacity planning, and performance analysis

REQUIRED QUALIFICATIONS

• Strong software engineering experience in Python and at least one systems language such as C++, C, HIP, CUDA, or Rust.
• Experience building applied AI, ML, agentic, automation, or developer tooling systems for technical users.
• Strong experience in at least one of the following:

• • Distributed training, reinforcement learning, or post-training systems
  • LLM inference, model serving, model optimization, or GPU performance
  • Ability to design and build reliable, scalable, and maintainable ML systems.
  • Strong debugging and root-cause analysis skills across software systems, AI workflows, or hardware-adjacent tooling.
  • Excellent collaboration and communication skills with domain experts, engineering leads, research teams, and program stakeholders.

PREFERRED EXPERIENCE

• Experience building or operating large-scale ML systems in production or research environments
• Experience with distributed computing, GPU clusters, workflow orchestration, or large-scale experimentation
• Familiarity with ROCm/HIP, CUDA, GPU profiling, or hardware-aware optimization
• Experience integrating ML systems with compilers, profilers, simulators, verification tools, or other engineering workflows
• Experience delivering platforms or systems used by multiple researchers or engineering teams
• Publications or shipped systems in ML systems, distributed training, reinforcement learning, inference, model optimization, or hardware-software co-design

EDUCATION
Bachelor's degree in Computer Science, Computer Engineering, Electrical Engineering, Machine Learning, or related field, or equivalent practical experience. Master's preferred; PhD is a plus, especially with work in ML systems, reinforcement learning, distributed systems, GPU computing, or AI infrastructure.

LOCATION: Santa Clara, CA

#LI-AG2

#LI-Hybrid

Benefits offered are described:  AMD benefits at a glance.

AMD does not accept unsolicited resumes from headhunters, recruitment agencies, or fee-based recruitment services. AMD and its subsidiaries are equal opportunity, inclusive employers and will consider all applicants without regard to age, ancestry, color, marital status, medical condition, mental or physical disability, national origin, race, religion, political and/or third-party affiliation, sex, pregnancy, sexual orientation, gender identity, military or veteran status, or any other characteristic protected by law.   We encourage applications from all qualified candidates and will accommodate applicants' needs under the respective laws throughout all stages of the recruitment and selection process.

AMD may use Artificial Intelligence to help screen, assess or select applicants for this position.  AMD's “Responsible AI Policy” is available here.

This posting is for an existing vacancy.