Researchers demo LLM agent running autonomous NV-center quantum sensing experiments

Researchers including Takuya Isogawa, Ryotaro Okabe, Nutdech Phadetsuwannukun, Mingda Li, and Paola Cappellaro describe an agentic AI workflow built around a large language model agent for autonomous experiments with nitrogen-vacancy centers in diamond, in a paper posted on arXiv as arXiv:2607.25145. NV centers are a widely used platform for quantum sensing, and computer-controlled measurements make them a natural setting for autonomous workflows. The authors demonstrate an autonomous NV experiment workflow that combines persistent project records, quantitative calculation and data analysis tools, and deterministic experiment control. In one autonomous experiment, the agent selected a single NV center, calibrated its resonant frequency, measured T2* with Ramsey measurements, and added a Carr-Purcell-Meiboom-Gill measurement to check a weak feature that could be related to nearby carbon-13. The team also introduced two offline benchmarks that evaluate the agent's reasoning separately from laboratory execution, tested with GPT-5.4, GPT-5.5, and GPT-5.6 Sol. In a Ramsey checkpoint benchmark, greater reasoning effort generally improved recognition of a residual resonance calibration offset. In a pulsed optically detected magnetic resonance data evaluation benchmark, pulse sequence information alone produced more false positive resonance judgments at higher reasoning effort, while requiring an expected signal calculation kept false positive rates low across models. The authors say the agent forms scientific hypotheses and uses quantitative tools to evaluate data, while deterministic code controls the hardware and enforces safety constraints.