PhD AI Research Infra Intern — RL Post-Training

Advanced Micro Devices

Santa Clara (CA)

Hybrid

USD 50,000 - 67,000

Full time

4 days ago

Be an early applicant

Application generator

Turn this role into an interview — a resume and cover letter built around what this employer wants.

Get past ATS filters

Job summary

Advanced Micro Devices in Santa Clara, CA is seeking a passionate AI Research intern who is currently pursuing a PhD in CS/ML/AI. You will work full time in a hybrid or onsite setup this Summer 2027, developing RL-based post-training infra for large models and building scalable tools for experimentation and deployment.

You should have strong Python and PyTorch skills, plus experience with RL, distributed training, and multi-GPU workloads.

Qualifications

Pursuing a PhD in CS/ML/AI or related field.
Strong Python and PyTorch programming skills.
Experience with RL, RLHF/RLAIF or preference optimization.
Experience with distributed training and large-scale inference.

Responsibilities

Develop and optimize RL-based post-training infrastructure for LLMs and multimodal models.
Build scalable rollout generation, inference, reward computation and policy updates.
Improve distributed training efficiency, reliability and memory usage.
Design interfaces for researchers to implement RL algorithms quickly.
Create tools for experiment config, logging, checkpointing and monitoring.
Profile end-to-end training pipelines and resolve bottlenecks.
Support on-policy and off-policy training workflows with verifiable feedback.
Document system designs and contribute to technical reports.