| 🏆 Signal | 📊 Proof Point |
|---|---|
| 7 Products Shipped | BreakoutScan · FirstStep · Regulatory Intel · FilingLens · Hyvemind · Career Seek · FlowKeys |
| Production Scale | 500+ concurrent users · 99.7% uptime · <2s API response · 30s data refresh |
| AI Architecture | Dual-model pipelines (Gemini + LLaMA) · RAG · Circuit-breaker failover · Evidence-cited AI |
| Fintech Depth | NSE-certified analyst · SEBI/RBI/IFSCA compliance · Retail investing · Options analytics |
| Solo Builder | Full-stack delivery without engineering team — Next.js, FastAPI, Redis, Supabase, Docker |
| Corporate PM | Ex-TCS · 12 PRDs · 5 verticals · 100% implementation fidelity · 45% miscommunication reduction |
| Community | 15+ FinGen Talks · Money Expo Summit 2025 speaker · GDG AI Solutions Challenge 2026 |
What this proves: I architect production-grade AI systems with real-time data pipelines and financial market domain expertise.
| ⚡ Speed | <2s NIFTY 500 scans · 30s price cache refresh · 3-8s AI Trade Brief generation |
| 🤖 AI Pipeline | Dual-model: Gemini 3.1 Pro (macro reasoning) + LLaMA 4 (pattern recognition) |
| 📡 Data Architecture | 5-layer broker API cascade — Groww → Upstox → Dhan → Fyers → Paytm Money with circuit-breaker failover |
| 🛠️ Stack | Next.js 15 · FastAPI · Redis · Supabase · Docker · Railway · Vercel |
| 📈 Impact | Retail traders save 60+ minutes of manual screening per market session |
What this proves: I design consumer fintech UX with behavioral economics, progressive disclosure, and India-first market understanding.
| 🧠 Product Design | 12-pillar guided investing — Trade Guard · Rupee Risk Slider · AI Coach (250 answers) · Jargon Buster · Goal Simulator |
| 🔓 Progressive UX | Zero real-money risk onboarding · Pro Mode unlocks after 5 learned terms |
| 🛠️ Stack | Next.js 15 · TypeScript · Tailwind CSS · shadcn/ui |
| 🇮🇳 India-First | INR pricing · Hindi context · SEBI-compliant education · No brokerage dependency |
What this proves: I build B2B/enterprise AI products with regulatory domain expertise and evidence-based AI architecture.
| 📋 Coverage | RBI · IFSCA · MCA · SEBI · FATF — daily circular monitoring with relevance scoring |
| 🤖 AI Design | Evidence-cited analysis — every claim grounded in verbatim document quotes with page-level citations |
| 🛡️ Resilience | Gemini 2.5 Flash → Groq LLaMA 3.3 automatic failover in <100ms |
| 🛠️ Stack | Next.js 15 · TypeScript · SQLite · Tesseract OCR · PDF parsing |
| 🏢 Origin | Production take-home for GIFT City IFSC-licensed fintech operator |
| Product | What It Proves | Key Signal |
|---|---|---|
| FilingLens | Complex RAG systems for enterprise compliance | 9 AI workflow templates · pgvector · SEBI/NSE/BSE filings |
| Hyvemind | AI media pipelines with multi-store architecture | Whisper → Gemini · 5-min setup · Clerk · Neon · R2 · Redis |
| Career Seek | Local-first productivity tools with multi-provider AI | 7 job boards · RAG Coach · SQLite · Meilisearch · BullMQ |
| FlowKeys | Native apps with India-specific product sense | macOS · Hindi/English/Hinglish · Keychain · No server |
| SoleMate | Cross-domain AI recommendation validation | 260 models · 16 brands · 94% match accuracy |
| Pattern | Implementation | Lives In |
|---|---|---|
| 🔁 Dual-Model Fallback | Gemini primary → Groq LLaMA secondary in <50ms | BreakoutScan, Regulatory Intel |
| 📜 Evidence-Cited AI | Every claim grounded in source with verbatim quotes | Regulatory Intel, FilingLens |
| 🔍 RAG Pipelines | pgvector + structured chunking + relevance scoring | FilingLens, Career Seek |
| ⚡ Circuit-Breaker Cascade | 5-layer API failover with auto-switching | BreakoutScan |
| 📡 Event-Driven Workflows | n8n · Redis pub/sub · WebSocket push | BreakoutScan, Bhukkad |
| 🔒 Local-First AI | Offline inference · privacy-preserving · no cloud | Career Seek, FlowKeys |
| Product | Metric | Value | Badge |
|---|---|---|---|
| BreakoutScan | NIFTY 500 scan time | <2 seconds | ⚡ |
| BreakoutScan | AI Trade Brief generation | 3–8 seconds | 🤖 |
| BreakoutScan | Price cache refresh | 30 seconds | 📡 |
| FirstStep | AI Coach answers | 250 India-first responses | 🇮🇳 |
| Regulatory Intel | Regulators monitored | 5 (RBI/IFSCA/MCA/SEBI/FATF) | 🏛️ |
| Regulatory Intel | AI fallback response | <100ms (Groq) | ⚡ |
| Hyvemind | 7-min video pipeline | 85 seconds | 🎬 |
| SoleMate | Match accuracy | 94% | 🎯 |
- Built 7 AI-native products across fintech, edtech, consumer, and productivity
- 500+ concurrent users on production systems with 99.7% uptime
- Consulted 8+ businesses on AI product strategy and roadmap prioritization (35% improvement)
- Delivered 15+ FinGen Talks on AI in finance to retail investors and students
July 2022 – August 2023 · Pune, India
- Developed 12 PRDs with quantified business metrics across 5 product verticals
- Led stakeholder alignment workshops ensuring 100% implementation fidelity
- Reduced cross-functional miscommunication by 45% via JTBD analysis
- Improved trial-to-paid conversion by 18% and time-to-first-action by 32%
- Delivered data-driven strategic positioning for 6 B2B SaaS clients via social listening
- 100% client adoption rate on recommended positioning strategies
- Co-published 8 features on fintech disruption and algorithmic trading
- Conducted 8+ expert interviews with financial analysts and market strategists
Frontend & Product
Backend & Infrastructure
AI / ML
Fintech & Data
Automation
Fintech · Consumer Tech · AI Infrastructure · EdTech · Regulatory Tech
Where I can own ambiguous problems, architect AI systems, and ship end-to-end.
