South Delhi, Delhi, India
I build systems that turn data into decisions. Over the past ~4 years, I’ve worked at the intersection of analytics, AI, and strategy, designing data-driven solutions that help organisations mitigate risk, elevate forecasting, and catalyse digital growth. Currently, I collaborate with cross-functional teams utilising large enterprise datasets to power commercial and product intelligence. My work focuses on analytical frameworks, model monitoring, and AI-driven capabilities that accelerate decision velocity and surface operational anomalies early. I recently architected a GenAI recommendation capability for a flagship platform driving enterprise revenue, boosting cross-sell performance by 33%. Previously at Incedo, I led the development of a GenAI-powered automated Data Quality platform, working with data engineers and product teams to improve how organisations validate and trust their data. The platform significantly reduced manual rule creation and accelerated product demo readiness by 33%. I hold a Bachelor of Engineering in Manufacturing Processes and Automation Engineering from NSIT(DU), New Delhi. Additionally, I have been actively involved in the non-profit sector, leading the admissions department of a school in a tier-3 town. I spearheaded initiatives such as the "Full Scholarship" program, which has facilitated the admission of 10-20 students annually from economically weaker sections (EWS), ensuring access to quality education for underprivileged communities.
everything product & AI
● Built an automated solution contextualizing Data Quality KPIs for domain-specific needs, integrating user feedback. Achieved a 90% efficiency boost and a 25% improvement in data accuracy, making the platform self-optimizing. ● Led Data Science team in developing pattern recognition, identifying data fingerprints and enhancing pattern recognition using Counter. Achieved a 50% increase in intelligent rule suggestions, leading to accurate insights. ● Enhanced complex data type profiling (e.g., US phone numbers, ZIP codes) using Python libraries and classes, reducing validation time by 60% and boosting data quality KPIs by 90%. ● Conducted in-depth bivariate analysis using Cramer's V (Cramer's V > 0.9), identifying critical one-to-one and one-to-many relationships to drive data insights, which led to a 33% increase in rule suggestions.