Laiba Sajjad completed Machine Learning for Automotive Safety

Built a perception validation dossier end to end for a synthetic Aurora-7 detector: defined IoU-based detection correctly, scored precision/recall and the VRU false-negative rate, sliced recall by night, occlusion, and rarity, and translated the worst misses into ISO 21448 SOTIF chains with concrete mitigations. The final release verdict was evidence-based and internally consistent, and the real Uber Tempe crash analysis correctly mapped to the night-time pedestrian failure mode.

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