3 years running real S&OP cycles in the UAE — now shipping AI products that fix the problems I lived every day.
Portfolio · LinkedIn — 7,000+ followers · Instagram · Email · WhatsApp · Resume
 |
 |
| Product | What it does | Stack |
|---|---|---|
| GCC Demand Forecasting Benchmark | Reproducible backtest showing Hijri-aware calendar regressors cut WAPE 13% and collapse forecast bias from +9.5% to −3.1%. Hierarchical reconciliation across 6 levels. 37 tests, 2 CI accuracy gates | Python · statsforecast · hierarchicalforecast |
| S&OP Decision Engine | Replenishment decisions computed from inventory policy — every number in the explanation traceable to the arithmetic, nothing self-approves. Wired to the forecaster above, it carries 20% less safety stock at the same service level. 61 tests, 8 golden decision cases gating CI | Python · FastAPI · SQLite |
| Meridian | CFA-level multi-market equity-research terminal (US · UK · UAE · Singapore) — live TradingView data + a Supabase RAG brain with an auditable multi-agent research pipeline | Supabase · pgvector · Deno Edge Functions |
| AlphaOS | AI-powered multi-market trading platform — TradingView charts, copy-trading & Claude-driven signals across US, India, UAE & Crypto markets | Next.js · TypeScript · Claude API |
| Project Autopilot | Supply-chain control-tower UI prototype — dashboards, simulator screens and a scripted decision feed. Front-end exploration, not a live decision system | Next.js · TypeScript · Supabase |
| Emirates Pride Inventory | iPad-optimised inventory system in daily use across Emirates Pride retail outlets in Dubai — GRN, stock transfers, sales & management reporting | HTML · JS · GitHub Pages (PWA) |
| S&OP Copilot | Runs Sunday night, delivers your Monday meeting brief before you wake up — AI briefing pack for UAE/GCC manufacturers | Claude API · Next.js · Supabase |
| GCC Event Surge Planner | Per-SKU demand forecasting for Ramadan, Eid, DSF, White Friday & National Day. The only tool that knows Ramadan moves 11 days a year | Claude API · Next.js · Supabase |
| AI Supply Chain Platform | OTIF Compliance Copilot + Exception Brain — two AI modules for supply-chain exception management | Python · Claude API · Supabase |
Languages: Python · TypeScript · SQL · DAX
AI/ML: Claude API · Prompt Engineering · Demand Forecasting (ARIMA, Holt-Winters, ML ensemble)
Frontend: Next.js · TailwindCSS · shadcn/ui · React · Framer Motion
Backend: Supabase (PostgreSQL + Auth + RLS) · REST APIs
BI Tools: Power BI · Excel (Advanced) · Google Sheets
Supply Chain: S&OP · Demand Planning · Procurement · OTIF · Inventory Optimization
Deploy: Vercel · GitHub Pages · GitHub Actions
- vinayakbhadani.github.io — Portfolio site — Demand Planning Analyst & AI Supply Chain Builder based in Dubai, UAE (updated 2026-08-05)
- gcc-simulator — GCC Business Simulator — Next.js 15 · Supabase · Google Gemini · 8 AI agents. by NayakLabs. (updated 2026-08-02)
- fuso-demand-dashboard — Demand planning dashboard for UAE commercial vehicle distributor — ARIMA, Holt-Winters, ML ensemble forecasting with interactive visualizations. (updated 2026-08-02)
- vinayak-portfolio-agent — Retrieval agent answering recruiter questions from source data. Local search runs with no API key. (updated 2026-08-02)
- gcc-retail-calendar — Announced Ramadan and Eid dates for the six GCC states, with provenance, as forecasting regressors. Astronomical conversion gets the day wrong 53% of the time. (updated 2026-08-02)
- gcc-demand-forecasting-benchmark — Reproducible demand-forecasting benchmark for GCC retail: Hijri-aware calendar regressors cut WAPE 14% and collapse forecast bias from +8.5% to -0.6% (updated 2026-08-02)
Actively targeting senior Demand Planning / Supply Chain roles across UAE's top FMCG, Retail, and Q-Commerce companies — and open to AI-product collaborations.
- Portfolio: vinayak682.github.io/vinayakbhadani.github.io
- LinkedIn: linkedin.com/in/vinayakbhadani
- Email: vinayakbhadani1998@gmail.com
- Location: Dubai, UAE
"I don't just plan supply chains — I build the tools that plan them better."