What Shubham demonstrated

Built an end-to-end rolling-bearing fault detector from vibration data: visualized a faulty snapshot, engineered RMS/kurtosis/crest-factor and envelope-spectrum defect-band features, identified an outer-race fault via BPFO, and trained a depth-capped decision tree on training bearings only. Evaluated on held-out bearings with reported catch-rate and false-alarm rate, then justified a low alarm threshold using explicit downtime-versus-inspection costs in a maintenance memo.

• Leakage-Free Feature Engineering
• Honest Held-Out Evaluation
• Cost-Based Thresholding
• Embedded-Budget Model Selection