Jairaj Narang
I build AI products where consumer culture meets data & finance.
Senior at Northeastern — Business Administration (Fintech) + Computer Science, Class of 2027. I've audited a Fortune 100's retail data at TJX, driven product analytics at OOFOS, and evaluated 50+ startups at Singh Capital Partners. Most recently, I built an AI that tracks sneaker demand — it updates itself every night.
SoleSight
hype, quantified.An end-to-end AI data product for the sneaker market: transformer sentiment on community chatter, ML demand forecasts, an LLM insight engine, and a fully autonomous pipeline that pulls fresh data nightly, recomputes every Hype Score, and redeploys itself — no human in the loop.
Internal Audit Co-op
Analyzed 1,000+ consumer transaction and merchandising records for a Fortune 100 retailer; built Power BI dashboards used by 10+ stakeholders across IT, Finance, and Risk.
Finance & Product Analytics Co-op
Product performance reporting for merchandising and FP&A; automated dashboards that cut reporting time by 50% and validated 100% data accuracy across 3 financial systems.
Investment Analyst Intern
Evaluated 50+ early-stage companies on product-market fit, TAM, and competitive positioning; synthesized findings into investment memos for senior leadership.
Social Media & Digital Analytics Intern
Drove data-driven audience growth across Reddit (16,625 new members), Instagram, LinkedIn, and Discord; built 20+ marketing assets and A/B-tested content strategy.
HuskyVision
GPS-free visual localization — dual-stream ResNeXt-50 CNNs with a POMDP framework, trained on 27,610 image pairs.
AI Fashion Classifier
CNN vs. FFN benchmark on 70,000 fashion images — 88% accuracy, 20+ experiments, insights for retail merchandising.
SPY vs. Crypto
Comparative dashboard of digital assets against traditional equity markets.
Birth Weight Prediction
ML regression on a 1,236-record maternal health dataset — KNN and Random Forest, feature importance analysis.
Off the clock: pickup basketball, sneaker archives, tabla riyaz, and NBA box scores — and a growing conviction that good products come from taste as much as data.