Lead Data Scientist

About the Company - Quartic.ai is a SaaS industrial AI start-up accelerating Industry 4.0 adoption by helping manufacturers unlock more value from existing infrastructure. Our flagship Quartic Platform empowers subject matter experts to build and deploy AI and IIoT applications—without needing programming or data science expertise. Used by Fortune 100 and 500 companies, the platform drives productivity and efficiency in sectors like pharmaceuticals, food & beverage, and CPG, supporting use cases such as process optimization, predictive maintenance, and energy management.

About the Role - Harness small industrial datasets to boost yield, quality, and efficiency. You will clean data, build lightweight ML models, and run Bayesian/black-box optimization loops—relying on deep mathematical and statistical understanding (PCA, PLS, Gaussian processes, etc.) to pick the right tool for each problem. No heavy software engineering; the focus is analytical rigor and measurable process impact.

Responsibilities -

• Data Quality + Exploration, audit, clean, and visualize sensor/log data quantify uncertainty Python (pandas, NumPy, Matplotlib), outlier & drift detection methods
• Statistical / Small-Data ML Modeling - regression, classification, clustering, PCA, PLS, Gaussian-process regression; rigorous CV and bootstrapping scikit-learn, XGBoost/LightGBM, PyMC/GPy, statsmodels
• Bayesian & Black-Box Optimization – tune process set-points or recipe parameters under real-world constraints; deliver closed-loop recommendations Optuna, Ax, BoTorch; expected-improvement, Thompson sampling
• Insight Delivery & Method Selection – map business questions to the correct statistical/ML technique and communicate results to engineers & executives Clear slide decks, concise memos, stakeholder workshops

Qualifications -

• Experience - 5+ years applying advanced statistical methods and machine learning to industrial processes or similar complex systems, with demonstrable impact on process optimization
• Education – Master’s in Statistics, Applied Mathematics, Industrial Engineering, or similar
• Mathematical Depth – linear algebra, probability, Bayesian inference, experimental error analysis
• Statistical Learning Mastery – PCA, PLS, RFE, LASSO/Ridge, Gaussian processes; ability to justify technique choice for each industrial scenario
• Optimization Experience – hands-on Bayesian or other black-box optimization applied to a real process (manufacturing, energy, materials, etc.)
• Programming Fluency – Python notebooks/scripts with pandas, NumPy, scikit-learn, and one gradient-boosting library
• Communication – translate quantitative findings into actionable operating guidance for mixed audiences

Preferred Skills -

• Probabilistic-programming exposure (PyMC, Stan)
• ML tracking/reproducibility tools (MLflow, DVC)
• Prior work in high-stakes industrial environments (chemicals, semiconductors, batteries, metals, etc.)