Repo Radar
AI-powered GitHub repository comparison tool that turns a 45-minute library evaluation into a 4-minute decision.
Content creators and marketers had no single tool to understand what was truly trending on YouTube — and why. YouTube Studio data lags by 24–48 hours, missing the critical first 6-hour window where algorithmic promotion is won or lost. Teams were spending 4–6 hours weekly manually stitching data from YouTube Studio, Google Trends, Social Blade, and Twitter into spreadsheets — with no cross-platform context and no natural language interface.
PULSE is a real-time YouTube analytics platform that aggregates trending data across 5 global markets every 5 minutes and wraps it in an AI chat layer powered by Groq LLaMA-3. Users can ask plain-English questions like ‘What gaming categories should I make a video about this week?’ and get structured answers with confidence scores and source attribution. Cross-platform signals from Twitter, TMDB, Spotify, and NewsAPI are overlaid on YouTube data — giving creators and marketers a single pane of glass for content intelligence.
I started by mapping the core user journeys for three personas: a solo creator optimizing for trend timing, a marketing lead reducing research overhead, and a media analyst looking for cross-platform signals. From there I prioritized a P0 feature set — live trending dashboard, AI chat, KPI analytics bar — and designed a graceful degradation architecture so the dashboard never goes blank on API failures. I ran a competitor analysis across Social Blade, TubeBuddy, VidIQ, and Brandwatch and identified the gap: no tool combined real-time YouTube data, AI natural language chat, AND cross-platform signals in a single free interface. The PRD defined a 4-phase roadmap from MVP to monetization, with Weekly Active Queriers as the north star metric.
Shipped a live product at pulse-eta-beryl.vercel.app with a full PRD, go-to-market strategy, and monetization model. The platform reduced manual content research effort by 60% for target users. Defined a freemium pricing model (Free / ₹499 Pro / ₹2499 Team) with a TAM of $2.5B and a 3-year SOM target of $50M in the AI-first YouTube intelligence category.
AI-powered GitHub repository comparison tool that turns a 45-minute library evaluation into a 4-minute decision.
Product concept for an AI-enabled loan underwriting agent for rural India — a voice-first, multilingual mobile app that enables credit access for new-to-credit borrowers using alternative data and explainable AI.
Mobile-first platform to collect geo-tagged, multilingual image-description pairs from every Indian village — building the largest India-centric vision AI dataset.
Product Management Portfolio