Dementia is an important challenge in an ageing society. Early detection is fundamental for timely intervention and improving the quality of life. The current diagnostic process of dementia is subjective and time-consuming, leading to delayed or missed diagnosis. We introduce a solution that, in an objective, fast and automatic way, can optimise the time-to-first assessment of cognitive impairment and enable large-scale screening. Our neurotech solution uses a proprietary stimulation protocol that elicits characteristic brain patterns (state-of-the-art biomarkers of dementia) from EEG signals and a proprietary data-driven AI algorithm for accurate, reliable, and explainable predictions, providing results that can be validated by a clinician. We have an MVP already tested in non-laboratory settings with promising results, and we are planning to validate it on a larger cohort, starting a pilot study. We aim to patent and deploy it in clinical settings.