Member of Technical Staff, Post-Training (AI)

Location: On-site in San Francisco or Tokyo

Company

hirify.global is an AI research lab developing generative genomics and biological foundation models to improve understanding, design, and treatment in biology while addressing associated biosecurity risks.

What you will do

• Develop and tune post-training recipes, datasets, reward signals, curricula, and training schedules for biological world models.
• Build evaluation suites for biological reasoning, scientific usefulness, long-context behavior, robustness, safety, and reliability.
• Debug training runs and model outputs end-to-end, tracing failures to data, optimization, evaluation, or systems issues.
• Explore preference modeling, reward modeling, synthetic feedback, rejection sampling, and related post-training methods.
• Define, curate, and generate expert-informed, synthetic, and task-specific post-training datasets.
• Study post-training scaling across dataset size, recipe complexity, compute budget, and model families while collaborating with training systems, architecture, and biology research teams.

Requirements

• Strong track record in ML research or engineering, particularly frontier-model training, post-training, alignment, evaluation, or data quality.
• Production-quality software and research infrastructure experience, ideally with Python and PyTorch, including large-scale training workflow debugging.
• Ability to design rigorous experiments, interpret ambiguous results, and distinguish real effects from artifacts, bugs, and benchmark overfitting.
• Excellent written and verbal communication across research, engineering, and scientific disciplines.
• Authorization to work in the United States is required.

Nice to have

• Experience with RLHF, RLAIF, preference optimization, reward modeling, rejection sampling, or other large-model post-training methods.
• Experience designing or operating evaluation frameworks for model quality, reliability, safety, or scientific performance.
• Familiarity with synthetic data generation, annotation workflows, or expert-in-the-loop data collection.
• Background in applied mathematics, systems, computational biology, or another quantitative scientific field.
• Contributions to open-source ML systems, model tooling, or research infrastructure.

Culture & Benefits

• Competitive compensation and comprehensive benefits.
• Support for continual learning.
• Collaborative, cross-disciplinary work spanning AI labs, biotechs, hospital systems, and research institutes.
• Research environment combining distributed systems, model architecture, numerics, and biological applications.