We use technical cookies and visit statistics: Yandex Metrica and Zorky's own de-identified visit log (cookie zk_aid). We never collect form contents, search text or full IP addresses, and never sell data. Opt out with the button below, or with Global Privacy Control or Do Not Track in your browser. More about cookies and statistics.
Zorky CRMZorky CRM
EN|RU
@termdocs
← All jobs

Principal Scientific Lead - Biophotonics & Data Analytics

principalhybridLondon, GBScore 59.5/1001d ago
Market insights
📊 AI / ML / Data Science: salaries and demand on the market
Stack
numpypythonpytorchscikit-learnscipy
Show company and link
Free. Sign in with email.
Apply
Upload your CV — we will connect you with the employer directly through our pool.
Send your CV →
Description
Join iLOF to lead the science that turns optical measurements of blood plasma into reliable, clinically useful biomarkers. Our Optomics® platform combines spectroscopy and AI to detect disease signatures in a blood sample, and we are moving from research towards validated tests.Our hardest problem sits at the interface between physics and data: extracting real biological signal from noisy, liquid-phase spectra (FTIR, Raman and related techniques) when there are thousands of spectral features but only a few hundred precious patient samples. Solving it well decides whether our models generalise to new patients, sites and instruments.As Principal Scientific Lead you will own that problem. You will set how iLoF acquires, cleans, models and validates spectral data, act as the senior scientific authority across our physics, data science, biology and hardware teams, and work directly with the Head of Innovation on iLoF's technical direction and longer-term technology bets. What you will own Signal-to-model methodology (about 40%)Set iLoF's standards for preprocessing, modelling and validation of spectral data in high-dimensional, small-sample settings (p ≫ n, typically N < 200 per cohort).Choose and develop the right methods: chemometrics and dimensionality reduction (e.g. PLS, PCA, wavelets, regularised and Bayesian models), and, where the data support it, physics-informed, self-supervised, metric-learning and transfer-learning approaches.Design data augmentation and domain-adaptat
Full description, company and the link to apply — after you sign in. Sign in
Employer contacts (email/phone/telegram) are hidden from the public preview — send your CV, and we will connect you directly.
Urgent question? Message @termdocs