
AI-focused platform contributor with 7+ years at Times Internet, driving 0→1 AI product enhancements across fintech and subscription ecosystems. Led incubation of ET Market GPT and AI Screener, transforming static financial tools into conversational intelligence systems covering 10,000+ companies. Experienced in PRD creation, backlog ownership, Natural Language → SQL workflow design, AI evaluation frameworks, and monetization-aligned product execution. Seeking Product Owner roles to formally lead AI-led platform strategy.
ETMARKET GPT: AI-Powered Conversational Financial Intelligence Platform
• Identified opportunity to transform static financial datasets into a conversational intelligence layer for retail investors.
• Defined product vision and user journey for ET Market GPT covering 10,000+ listed companies and indices.
• Conducted competitive research across 1,500+ external screeners to identify high-frequency financial attributes used by investors.
• Rationalized 4,000+ raw financial fields into ~1,000 high-impact attributes, reducing AI dataset exposure by ~70% to optimize token cost and response efficiency.
• Designed end-to-end Natural Language → Structured SQL workflow:
User Query → Query Enhancement Layer → SQL Generation → Data Fetch → Tabular Output → AI Interpretation
• Authored detailed PRDs and JIRA epics, created user stories with acceptance criteria, and collaborated with design (Figma) for conversational UX flows.
• Partnered with engineering to implement vector similarity-based schema mapping for financial field normalization.
• Defined AI evaluation framework using percentile scoring and reasoning analysis to benchmark GPT, Claude, and internal models.
• Introduced AI credit governance framework (daily/monthly limits) to align LLM usage with subscription monetization strategy.
AI SCREENER: AI-Enhanced Stock Screener (Manual → AI Mode Transformation)
Earlier State:
Manual screener requiring structured financial query input (e.g., Market Cap > 500 Cr).
My Innovation:
• Identified friction in manual screener experience for non-technical retail investors.
• Conceptualized and launched AI Mode for Screener:
Enabled users to type natural language queries instead of structured financial syntax.
• Designed backend workflow to:
Natural Language Input → Intent Detection → Financial Condition Mapping → SQL Query Generation → Screener Output
• Ensured users could:
• Defined product requirements, created JIRA user stories, aligned UX with design team, and delivered KT to engineering & QA teams.
• Enhanced product accessibility for non-finance-savvy users, increasing engagement potential across broader retail audience.
Ownership:
TECHNICAL & AI EXPOSURE: