This role pays less than
75%
of similar roles. Most pay
$114,437–$246,125
— the shaded band above.
At the midpoint, this role pays about
$114k
versus about
$180k
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
474 open roles
on FindRole.
Listed pay typically runs
$166,400–$249,600
across 474 roles with salary data.
The Summer 2027 PhD Gen AI and Reinforcement Learning Research Intern will join the research team to focus on post-training methods for large generative models. The role involves researching and prototyping techniques for policy optimization, preference learning, reward modeling, exploration, and credit assignment. You will develop methods to improve reasoning, code generation, tool use, and agentic behavior while designing controlled experiments using verifiable or model-generated feedback. Key responsibilities include analyzing failure modes like reward hacking and training instability, as well as developing evaluations for model safety and robustness. The position requires expertise in reinforcement learning, deep-learning methods, Python, and PyTorch. Candidates should have experience with LLM post-training, RLHF/RLAIF, and multimodal agents. This role addresses technical challenges in generative AI, reasoning models, and large-scale training within the broader AI ecosystem.
Research and prototype methods for post-training large generative models.
Explore policy optimization, preference learning, reward modeling, and credit-assignment techniques.
Develop methods to improve reasoning, code generation, tool use, and agentic behavior.
Design and run controlled experiments using verifiable or model-generated feedback.
Analyze failure modes such as reward hacking, policy degeneration, and training instability.
Create evaluations to measure model quality, robustness, safety, and real-world performance.
Document findings and contribute to technical reports and publications.
What we're looking for
Must be currently pursuing a PhD in Computer Science, Machine Learning, Artificial Intelligence, Electrical Engineering, Computer Engineering, or a related field.
Strong knowledge of reinforcement learning and modern deep-learning methods.
Experience implementing and evaluating machine-learning models using Python and PyTorch.
Familiarity with LLM post-training, RLHF/RLAIF, preference optimization, or language and multimodal agents.
Experience conducting reproducible experiments and analyzing empirical results.
Publications at leading machine-learning or computer-vision conferences are preferred (preferred).
Experience with large-scale model training, GPU computing, or distributed systems is beneficial (preferred).
Familiarity with generative AI, multimodal learning, reasoning models, code models, or agentic systems is a plus (preferred).
Machine Learning
Generative AI
Reinforcement Learning
LLM Post-training
RLHF
ML Systems
AI Agents
Reward Modeling
Preference Optimization
GPU Capacity Planning