Frontier AI Research Scientist

Turing; United States; On-site; USD 150,000 - 300,000; Full time

Job summary

Turing is seeking exceptional AI Research Scientists to join our STEM research organization and develop new benchmarks, datasets, and evaluation methods for frontier AI systems. This research-first role focuses on identifying gaps in literature, proposing ambitious directions, and guiding projects from hypothesis to publication.

We focus on frontier STEM evaluation, synthetic data, and agentic science, seeking scientists who can define problems precisely and execute rigorous programs to produce

Qualifications

• PhD or equivalent research experience in a highly technical field such as ML, CS, mathematics, physics, chemistry, biology, engineering, statistics.
• Demonstrated ability to formulate and execute original research.
• Strong understanding of modern LLMs and frontier AI research landscape.
• Excellent experimental design, quantitative reasoning, and scientific judgment.
• Ability to rapidly understand unfamiliar technical literature and develop expertise in new areas.

Responsibilities

• Identify high-impact gaps in existing benchmark and evaluation literature.
• Design novel benchmarks in STEM fields and general model functionality.
• Develop evaluations for emerging model capabilities not well captured by static benchmarks.
• Design rigorous task-generation, grading, contamination-control, difficulty-calibration, and validation methodologies.
• Build benchmarks that become valuable research contributions and standards for frontier models.
• Develop methods for generating high-quality synthetic STEM training data.
• Study how data quality, difficulty, and verification affect downstream performance.
• Investigate how synthetic data can improve capabilities without simply increasing training volume.
• Study hallucination, uncertainty, calibration, and epistemic failure in technical domains.
• Develop evaluations and methods for improving factual reliability and verification.
• Research agentic science workflows involving literature search, coding, simulation, tool use, and iterative reasoning.

Skills

Python programming

Experimental design

Scientific writing

LLM knowledge

Research methodology

Education

PhD or equivalent

Job description

Turing is seeking exceptional AI Research Scientists to join our STEM research organization and develop new benchmarks, datasets, and evaluation methods for frontier AI systems. This research-first role focuses on identifying gaps in literature, proposing ambitious directions, and guiding projects from hypothesis to publication.

We focus on frontier STEM evaluation, synthetic data, and agentic science, seeking scientists who can define problems precisely and execute rigorous programs to produce