Alex Binuraj completed Machine Learning for Automotive Safety

Built a perception validation dossier that correctly defined IoU-based detection matching, scored a synthetic detections table, and identified VRU false-negative rate as the safety-critical metric. The work goes beyond aggregate accuracy by slicing recall across night, occlusion, and rarity, then translating the worst misses into concrete SOTIF mitigations such as sensor-fusion fallback and operating-domain restrictions.

• Leakage-Free Detection Definition
• Honest Safety Metric Selection
• Per-Condition Failure Analysis
• SOTIF Functional-Insufficiency Reasoning