Applied AI Researcher — Post-Training RL & Datasets

True Talent Solutions Partners; New York (NY); Ibrido; USD 300.000 - 500.000; Tempo pieno

Vantaggi offerti da questo lavoro

Visa sponsorship available

Descrizione del lavoro

True Talent Solutions Partners is seeking an experienced applied researcher to work with frontier AI research labs on post-training problems. You will translate open-ended research goals into experiments and training data to improve model capabilities.

The role involves designing RL loops, building evaluation harnesses, and analyzing results across distributed GPU clusters to guide future research in a hybrid New York environment, with Seattle as an additional listed location.

Competenze

• 2+ years as an applied researcher or research scientist with hands-on model training experience.
• Practical reinforcement learning and agent experience beyond standard ML engineering work.
• Research experience at frontier AI labs, emerging labs, or notable publications.
• Strong understanding of reinforcement learning and training data with demonstrated outcomes.
• Proficiency in Python, PyTorch, FSDP, FastAPI, GRPO, and distributed GPU clusters.

Mansioni

• Partner with customer research teams to translate goals into concrete environments, datasets, experiments, and deliverables.
• Design and run post-training loops using reinforcement learning methods such as GRPO, preference optimization, and supervised fine-tuning.
• Evaluate training signals that improve model performance and shape high-fidelity trajectories and datasets with attention to edge cases.
• Build reward functions, rubrics, validators, and evaluation harnesses that resist reward hacking.
• Analyze post-training signals and experiment results across distributed GPU clusters to guide subsequent research.

Conoscenze

Applied researcher

Reinforcement learning

Python

Distributed GPU clusters

Research publications

Strumenti

PyTorch

GRPO

FastAPI

FSDP

Descrizione del lavoro

True Talent Solutions Partners is seeking an experienced applied researcher to work with frontier AI research labs on post-training problems. You will translate open-ended research goals into experiments and training data to improve model capabilities.

The role involves designing RL loops, building evaluation harnesses, and analyzing results across distributed GPU clusters to guide future research in a hybrid New York environment, with Seattle as an additional listed location.