Senior AI Product Manager · San Francisco, CA
AI-focused Product Manager with 6+ years of experience turning customer problems into 0→1 AI and data products, spanning agentic AI, multimodal RAG, and personalized recommendation systems. I combine customer discovery, product strategy, experimentation, and hands-on technical execution to launch and scale products.
I also build what I spec. The projects below are working, open-source tools.
| Project | What it does | Stack |
|---|---|---|
| shipgate | Go/no-go eval harness that scores an AI feature against a quality bar and returns GO / NO-GO with a CI exit code. | Evals, CI |
| Enterprise-AI-Chat-Agent | AI agent platform with RAG, a file search API, and configurable multi-agent creation for document analysis and content generation. | TypeScript |
| murmur | News-grounded forecasts by Claude, with a one-million-person reaction simulation that runs in the browser. | JavaScript |
| Lookout-Assist | Multimodal "point-and-understand" assistant for unfamiliar objects, control panels, foreign-language signs, and software crashes. | TypeScript |
| evoo | Free voice dictation for Mac that runs 100% on-device. Hold fn, speak, release. | Swift |
| Ad-Intel-LinkedIn- | Crawls the LinkedIn Ad Library, enriches creatives with Gemini, and generates brand and competitor dashboards. | TypeScript |
| Project | What it does | Stack |
|---|---|---|
| facet | Open-source responsive design tester: see a site on 290+ phones, tablets and desktops at once, with layout audits on each. Has a web version, and a CLI and MCP server so AI coding agents can check pages. | TypeScript, Electron |
| Project | What it does | Stack |
|---|---|---|
| ledgerline | Daily rule-based buy / exit / sell signals for liquid US and NSE stocks, with an open method, a backtest, and a public append-only signal log. | Python, GitHub Actions |
| pigeonai | Open-source NSE stock scanner and charting site with 53 scans on split/bonus-adjusted end-of-day data. | TypeScript |
| pm-jobs-usa | Product Management job listings across 60+ companies, refreshed automatically every 24 hours. | Python |
Enterprise SaaS · San Francisco, CA · Dec 2023 – Present
- Own the AI product portfolio for an enterprise work-management platform serving 100+ enterprise customers managing $10B+ in annual project value; launched packaged AI tiers that drove $3M in incremental ARR.
- Led 0→1 development of multi-agent AI solutions that synthesize documents, images, chats, forms, and tables into grounded, actionable outputs. Designed autonomy boundaries, tool permissions, and confidence-based escalation, achieving >80% task completion with near-zero critical incidents.
- Increased AI answer acceptance from 61% to 84% by turning customer feedback and failure patterns into evaluation requirements: golden datasets, human-calibrated LLM-as-judge, regression gates, and human-in-the-loop review.
- Optimized quality, latency, and inference cost through model selection, complexity-based routing, and caching; established quality SLOs and cost-per-task economics for deployment across 100+ enterprise tenants.
Social Media · Bangalore, India · Apr 2021 – Dec 2023
- Conceptualized the product strategy and launch of a Twitter-like social network for the trading community through 100+ customer interviews, competitive and regulatory research, and a beta launch to validate product-market fit.
- Led the transition from a chronological to an algorithmic personalized feed, defining ranking objectives and running ranking experiments with data science; increased session duration by 15% and retention by 12%.
- Designed the creator growth roadmap with incentive and reputation mechanisms, increasing creator participation by 31%.
- Championed continuous A/B testing and iterative UX updates such as media previews and thread nesting, driving a 22% increase in user engagement.
Quantitative proprietary trading · Bangalore, India · Jul 2017 – Mar 2021
- Analyzed trader interaction data on terminals and streamlined order-entry workflows to reduce execution delays and manual errors.
- Optimized the post-trade analysis workflow, accelerating the feedback loop for quants to test, iterate, and deploy new trading strategies.
| Degree | Institution | Year |
|---|---|---|
| MS, Business Analytics | Foster School of Business, University of Washington | Expected 2027 |
| Professional Degree, Product Management | Indian School of Business, Hyderabad | 2023 |
| BSc, Economics & Mathematics | University of Delhi | 2017 |
- Product: Roadmapping, Discovery, 0→1, Scaling, A/B Testing, Experimentation, UX Research, Market Research, Competitive Analysis, Stakeholder Management, Go-to-Market, MVP, Agile
- AI: LLM Agents, Guardrails, Fallback Strategies, Model Evaluation (Evals), Conversational AI, RAG Architecture, Prompt Engineering, Automatic Speech Recognition, ETL Architecture, LangChain
- Tools: Cursor, Claude Code, Gemini, n8n, LangSmith, LibreChat, Figma, Tableau, SQL, Python, GitHub, Jira, Notion
The best way to reach me is on LinkedIn.