Mohak Shah completed Machine Learning for Automotive Safety

Built a perception validation dossier that correctly defined detection at IoU >= 0.5, scored the synthetic Aurora-7 table, and separated headline recall from the safety-critical VRU false-negative rate. Then sliced recall by lighting, visibility, and rarity to surface the night/occluded/rare VRU failures, framed them as SOTIF functional insufficiencies, and issued a conditions-based release verdict grounded in the dossier’s own numbers.

• Leakage-Free Detection Definition
• Safety-Critical Metric Selection
• Per-Condition Failure Triage
• SOTIF Hazard Framing