Vansh completed Machine Learning for Automotive Safety

Built a complete perception-validation dossier for a synthetic Aurora-7 detector: defined detection at IoU ≥ 0.5, scored precision/recall and VRU false-negative rate from the table, and sliced recall by lighting, visibility, and rarity to expose the night and occlusion failure modes. The final recommendation was evidence-based and internally consistent, with SOTIF mitigations and a Tempe-crash comparison tied back to the same safety-critical perception gap.

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