parmeet singh completed Machine Learning for Automotive Safety

Built a perception validation dossier for a synthetic Aurora-7 detector that defined detection correctly at IoU ≥ 0.5, computed aggregate and VRU-specific precision/recall from a scored table, and then sliced recall by lighting, visibility, and rarity to expose the dangerous night and occlusion failures. The final release recommendation was evidence-based and internally consistent, with SOTIF mitigations tied directly to the highest-risk vulnerable-road-user misses and a real-world comparison to the Uber ATG Tempe crash.

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