Case Studies
Real tests. Real data.
See exactly how our AI performs on real-world datasets, side by side with the competition.
HiringML vs GPTCase StudyBenchmarks
We Tested Two AI Systems on the Same 16 Candidates
A head-to-head test: our proprietary ML model vs OpenAI GPT, both given identical hiring data. 87.5% agreement, 259× speed difference, $0 vs $6 per 1,000 runs.
May 20254 min read
Read case studyTradingPattern ModeRecommenderCase Study
When ML Signal Is Too Low, Lensiq Switches to Pattern Mode Automatically
A real trading use case: ML confidence at 67.2% triggered Pattern Mode, surfacing per-segment RR averages instead of a point prediction. See how the model chose transparency over false precision.
June 20253 min read
Read case studyLogisticsDeliveryCase StudyLebanon
Lensiq predicted 99 last-mile courier deliveries in Lebanon with 92.93% accuracy
A real-world delivery prediction test on last-mile courier and e-commerce fulfillment data in Lebanon. 92 of 99 correct, 94.2% average confidence, and the model flagged its own uncertainty on almost every mistake.
May 20255 min read
Read case study