QuantorSignal ranks a stock universe on the factor you care about, then shows what clears the bar. Switch factors, drag the threshold — the board re-ranks and filters live, just like the real screener.
Rank by factor—
Threshold
Score ≥70
A live re-creation. The product runs the real factor model over the S&P universe with a backtester behind it.
A 256-channel net records the scalp; spatial filtering and a trained classifier turn micro-volts into a decision. Pick a classifier, choose the stimulus, and fire — watch the response ripple across the montage and get decoded.
256-channel montageidle
Classifier
Stimulus
Live channels & decodestreaming
—
awaiting stimulus0%
Illustrative of the decoding pipeline (target vs. non-target, d′ & AUC scored). Not live subject data.
03 / BIOLOGYT2PW · research internship
Turning a paragraph of biology into a living pathway.
T2PW reads plain-language pathway descriptions and rebuilds them as structured, ID-mapped diagrams — extraction, enrichment, and grounding to HMDB / KEGG / ChEBI. Extract the pathway, then reveal the database IDs it resolved.
Input · plain textidle
Glucose is phosphorylated by hexokinase to form glucose-6-phosphate, consuming one ATP and releasing ADP. The glucose-6-phosphate is then isomerized to fructose-6-phosphate by phosphoglucose isomerase.
Using MIMIC-III intensive-care records, this study measures how often disease-prediction models miss real cases — the false-negative rate — and whether those misses land unevenly across demographic groups. Then it tests a reweighting fix across five conditions, with every change significance-tested. The honest answer is mixed: it helps in some cohorts more than others. Presented July 2026.
