ML Engineer (US Market Research SaaS)

Digital Quick is a Market Research SaaS company seeking an ML Engineer to build and maintain production machine learning models for survey respondent certainty prediction and fraud detection using behavioral signals.

As the sole ML engineer, you'll own hybrid CNN+tabular models and develop bot detection classifiers while working directly with a US-based AI scientist.

The role focuses on classical ML and deep learning applied to structured behavioral data in a production environment.

• Build and own ML models that predict survey respondent certainty and detect fraud from behavioral signals including mouse movement, click patterns, timing, and browser/OS data
• Improve the hybrid 1D CNN (mouse trajectories) + MLP (tabular features) certainty model through error analysis, hyperparameter tuning, augmentation, and loss experiments
• Build the bot detection classifier as a binary human/bot model reusing behavioral pipeline infrastructure
• Run segment-level evaluations with per-survey quality checks, precision/recall optimization, and calibration analysis
• Track experiments in MLflow with every run logged including parameters, metrics, and dataset versions
• Process raw time-series data converting mouse trajectories into [T, 7] tensors and document work to audit standards

• Strong PyTorch experience with ability to build and train CNNs from scratch, not just fine-tuning
• Tabular ML experience with LightGBM, XGBoost, or gradient boosted trees
• Rigorous evaluation mindset including calibration, segment analysis, label quality assessment beyond single-metric optimization
• Time-series/sequence data experience identifying temporal patterns in any domain
• Imbalanced classification expertise with precision/recall tradeoffs, threshold tuning, and AUC-ROC understanding
• MLflow or similar experiment tracking tool experience (W&B, Neptune)
• Self-directed work approach as the only ML person on execution with strategic guidance from US AI scientist
• Ability to produce auditable work with clean notebooks and scripts reviewed by advisors
• Clean verbal English communication skills

Schedule & Location: Fully remote work from anywhere with reliable overlap to CET timezone

Benefits & Perks: 20 paid days-off per year, 10 US/UA holidays, flexible schedule, minimal bureaucracy, high autonomy and trust, fast decision-making, senior team members and strong AI Scientist Advisor support