Piyush Thakur completed Machine Learning for Automotive Safety

Built a perception validation dossier end to end for a synthetic Aurora-7 detector: defined detection by IoU>=0.5, computed aggregate precision/recall and VRU false-negative rate, and exposed the dangerous night/occlusion slices with per-condition recall analysis. The final recommendation was evidence-based and internally consistent, pairing a daylight-only release with concrete mitigations and a correct comparison to the Uber Tempe crash.

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