THE ROLE:
We are hiring a Technical Program Manager, AI Research to drive execution across the AI Research scientists and research infrastructure engineers advancing reinforcement learning and post training, generalization across hardware engineering contexts, and the RL infrastructure that makes large scale experiments productive.
You will help research leaders turn scientific agendas into executable plans with clear milestones, resource needs, evaluation standards, and landing criteria. You coordinate across research workstreams, engineering partners, platform teams, and hardware domain experts so that experiments are reproducible; results are interpretable, and promising methods have a credible path from ablation to broader internal use.
THE PERSON:
The ideal candidate thrives in fast-paced, research-driven environments and can translate complex AI research into clear plans, risks, and priorities. They are proactive, organized, and skilled at aligning cross-functional teams, resolving blockers, and moving programs from experimentation to broader impact.
- Excellent communication skills with research scientists, engineers, directors, and executive stakeholders.
- Strong program discipline: milestone tracking, dependency management, decision records, and concise status synthesis.
Teams and Programs in Scope
The candidate will own one or multiple of the programs listed below.
- Research Scientists, Reinforcement Learning and Post training: policy optimization, preference based methods, reward modeling, stability, and scaling.
- Research Scientists, Hardware: transfer, metareasoning, cross program benchmarks, and learning under slow or heterogeneous silicon signoff feedback.
- Research Scientists / Engineers, Infrastructure for RL: distributed training, rollout generation, trajectory stores, researcher facing APIs, and reliability.
KEY RESPONSIBILITY:
- Own research program planning: quarterly and annual milestones, experiment roadmaps, publication or technical report timelines, and dependency tracking across research pillars.
- Coordinate research execution with multiple AI teams on datasets, eval harnesses, compute allocation, and method landing criteria.
- Maintain research operating cadence: milestone reviews, experiment readouts, decision logs, and escalation of scientific or infrastructure blockers.
- Track progress against research goals using clear metrics: benchmark movement, ablation completeness, reproducibility, stability, generalization evidence, and safety or governance checkpoints for RSI pilots.
- Manage compute planning and experiment throughput: GPU job scheduling assumptions, cluster capacity negotiations, and prioritization across competing research threads.
- Facilitate cross pillar collaboration, especially where RL, hardware generalization, and RSI intersect on reward design, evaluation, and containment.
- Prepare executive and leadership updates that explain scientific progress, uncertainty, risks, and recommended investment shifts in plain language.
- Help define promotion paths from research pilots to engineering adoption, including documentation of interfaces, eval definitions, and rollback criteria.
Technical focus areas
- Familiarity with Reinforcement learning and post training for LLMs, code generation models, and agentic workflows.
- Benchmark design and process to evaluate AI models on hardware engineering tasks.
- Familiarity with Large scale infrastructure: distributed training, rollout workers, logging, checkpointing.
PREFERRED EXPERIENCE:
- Experience managing ambiguous, experiment driven programs.
- Technical program management, research program management, or equivalent experience in AI, ML, or large scale compute organizations.
- Fluency in machine learning, reinforcement learning, LLM post training, or ML systems, enough to follow research plans, surface schedule risk, and coordinate across scientists and infra engineers.
- Experience supporting AI research labs, and ML R&D organizations.
- Familiarity with publication cycles, internal tech report processes, and research into engineering handoff practices.
- Exposure to RLHF, GRPO, preference optimization, reward modeling, or large-scale experiment management.
- Experience with GPU cluster planning and distributed training workflows.
- Familiarity with hardware design, verification, EDA, or hardware/software co-design environments.
ACADEMIC CREDENTIALS:
- Bachelor's degree in Computer Science, Computer Engineering, Electrical Engineering, Machine Learning, or related field, or equivalent practical experience is preferred.
#LI-BM1
#LI-Hybrid
Benefits offered are described: AMD benefits at a glance.
