Associate Product Manager with experience building AI-powered agents, conducting data-driven product teardowns of consumer platforms, and defining success metrics for AI-first products. Strong in problem framing, writing PRDs, and feature prioritization.
AppsForBharat
Identified content drop-off across 250+ temple partners; designed a structured coordination workflow that improved daily upload consistency- directly impacting retention on the Daily Darshan app. Scoped and shipped an AI-powered photo darshan pipeline (Google Flow) to solve low-quality image uploads from temples , reducing manual processing and enabling a scalable content standard across 250+ partners. Reduced manual photo processing effort significantly by automating a previously human-driven workflow, improving content throughput and consistency for the Daily Darshan app.
Honasa Consumer (Mamaearth)
Led cross-functional coordination between operations, analytics, and product teams to streamline workflows and improve operational efficiency across customer-facing processes. Built data-driven insights using SQL, Excel, and Python to analyze user behavior, identify bottlenecks, and support product decision-making. Worked extensively with dashboards and BI tools to track KPIs, monitor performance trends, and support business reporting initiatives.
Lexconn
Synthesised feedback from 1,000+ users into structured PRDs, cutting requirement ambiguity and enabling the team to ship 2 features ahead of schedule. Drove feature prioritisation using the RICE framework, converting user pain points into actionable PRDs and user stories that informed sprint planning and roadmap decisions. Partnered with stakeholders to define success metrics and OKRs for key initiatives, ensuring measurable outcomes were tracked and communicated to leadership.
B.E.
Relevant Courses: Business Analytics, Database Management Systems, Data Privacy & Ethics, Data Structures
Designed and built an AI agent to aggregate PM roles from LinkedIn, Indeed, Internshala, and Naukri.com — automating daily discovery, filtering, and Google Sheets updates (company, JD, required skills, apply link). Implemented RAG-based JD parsing using chunking strategies and vector embeddings for semantic relevance matching, significantly outperforming keyword search. Achieved measurable impact: 120–150 relevant roles identified weekly with<5% noise, saving∼3 hours/week of manual job searching.
Reverse-engineered Spotify’s recommendation engine, mapping 3 user personas (passive listener, commuter, nostalgic), Jobs-to-be-Done frameworks, and the full user journey across Music and Podcasts surfaces. Analyzed system design across content-based filtering, collaborative filtering, and knowledge-based approaches — including audio signal analysis (energy, valence, danceability), metadata, and perturbation techniques. Defined a success metrics framework with North Star metrics and L1 input metrics for both implicit feedback (saves, playlist adds, shares) and explicit feedback (time spent on recommended tracks) loops.
Coursera / Google
Saturnalia, TIET
Thapar Movie Club
Led 80+ member team; executed CineYouth for 3,000+ attendees managing budgeting, logistics, and end-to-end delivery.