SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
iLoF is building an AI-powered diagnostic platform that uses optical fingerprinting of blood plasma to enable personalized medicine. The company has developed Optomics®, a proprietary platform combining spectroscopy and machine learning to detect disease biomarkers, with current focus on Alzheimer's detection alongside verticals in digestive cancer, stroke, and infectious diseases. Backed by Microsoft Ventures, Mayfield, and Oxford University, iLoF is transitioning from research to validated clinical tests.
As Principal Scientific Lead, you will own the scientific core of the platform, sitting at the critical interface between physics and data science. You will report directly to the Head of Innovation and operate with high autonomy to shape the technology roadmap and make scientific decisions within your remit.
Your responsibilities span three main areas:
**Signal-to-Model Methodology (40%):** Set iLoF's standards for preprocessing, modeling, and validation of spectral data in high-dimensional, small-sample settings (p ≫ n, typically N < 200 per cohort). Select and develop appropriate methods including chemometrics, dimensionality reduction (PLS, PCA, wavelets), regularized and Bayesian models, and where data support it, physics-informed, self-supervised, metric-learning, and transfer-learning approaches. Design data augmentation and domain-adaptation strategies using physically grounded synthetic spectra to expand training data without bias or leakage. Determine when models are scientifically ready to advance by reviewing validation, robustness across batches/sites/instruments, and overfitting risk.
**Spectral Interpretation and Experimental Design (40%):** Interpret raw optical signatures from plasma and nanoparticle samples; diagnose artifacts such as baseline drift, autofluorescence, scattering, and instrument variability. Act as the technical bridge between hardware, data science, and biology teams—checking physical variables (integration time, laser stability, detector noise) before proposing algorithmic solutions. Design experiments and lab protocols to maximize information content and efficient use of scarce clinical samples. Mentor scientists and data scientists on data physics and rigorous experimental practice.
**External Scientific Partnerships (20%):** Lead scientific relationships with CROs, contractors, and academic collaborators. Write technical statements of work, set data and code standards, and review deliverables. Work peer-to-peer with university principal investigators to keep collaborations aligned with iLoF timelines. Ensure data and code shared externally protects iLoF's intellectual property in coordination with the data protection lead.
The role is designed to grow into technical executive leadership. You will work in a London-based team with offices in Porto, in a flexible, remote-friendly environment with a strong emphasis on collaboration, ownership, and scientific rigor.
**Requirements:**
**Must Have:**
- PhD in biophotonics, computational physics, chemometrics, applied mathematics, or closely related quantitative field.
- One of the following:
- Industry/translational route: 5–8+ years post-PhD delivering R&D outcomes in industry, deep-tech start-up/scale-up, scientific consultancy, or translational research, including leading technical work through to a product, validated method, or clinical study; OR
- Research route: 8+ years post-PhD research with strong publication record in spectroscopy, chemometrics, biomedical signal analysis, or ML for physical data, plus track record leading major projects (e.g., principal/co-investigator on large funded programs or multi-partner consortia).
- Proven track record building predictive models on real instrument-generated data (ideally spectral or optical) with tightly constrained sample sizes. Experience limited to simulated or standard benchmark datasets is insufficient.
- Deep expertise in chemometrics and statistical learning for high-dimensional, correlated data, including rigorous validation and prevention of data leakage.
- Strong Python proficiency (NumPy/SciPy, scikit-learn, PyTorch) with emphasis on transparent, reproducible analysis.
- Experience setting scientific direction for others and managing external scientific partners or contractors.
**Advantageous:**
- Experience with blood plasma or other biofluid spectroscopy.
- Physics-informed ML, self-supervised learning, metric learning, or domain adaptation.
- Exposure to diagnostics, IVD, or regulated product development.
