FA.

AI-SaaS · NEXA Tech Labs

NEXA-Sphere

Prototype

A B2B AI-SaaS platform concept for optimizing inventory and supply-chain operations of multi-outlet F&B businesses, built around demand forecasting, procurement recommendations, and a centralized dashboard.

Problem

Multi-outlet F&B SMEs (2–10 branches) are too small for enterprise ERP yet too complex for manual, instinct-based purchasing — leading to overstock, stockout, and food waste.

Solution

A B2B AI-SaaS platform that turns historical point-of-sale data into purchasing recommendations: demand forecasting, procurement recommendations with owner approval, and a centralized multi-outlet dashboard.

My role

Team Lead & AI/ML Engineer

Contribution focus

Technology architecture direction, predictive-model (demand forecasting) development, and algorithm risk management; presented the proposal at ICBC 2026 as team lead.

Technology stack

  • Time-series forecasting (ARIMA, Prophet) — proposed
  • Gradient boosting (XGBoost / LightGBM) — proposed
  • Cloud (AWS / GCP) — planned

Current status

Business-proposal concept with interface prototypes — finalist (Top 4, Juara Harapan 1) at ICBC 2026. Not released and not yet in wide use. Screenshots in the proposal use simulated internal data.

Supports skills

  • AI Engineering & Data Science
  • Product Management
  • Business & Entrepreneurship

Core pillars (proposed) Proposal

  • AI demand forecasting (daily / weekly sales per menu item)
  • Procurement recommendations with owner approve-or-modify control
  • Centralized dashboard across outlets (mobile and desktop)

Prototype shown in the proposal (simulated data) Prototype

  • POS Terminal with an AI Engine Analytics view
  • Floor Plan Manager and Admin dashboard (Dynamic Pricing toggle, Live Metrics)
  • Prototype navigation tabs: Kitchen Display, Table Map, Data Science HQ, Inventory, Live Chat, Public Web
  • Dimsum Manager with design partner Dimsum Mentai Kemayoran — role-based login (Owner / Admin / Kasir) live; about 80% complete at the time of the proposal

Planned, not yet built Planned

  • Integration with local POS providers (Moka, Pawoon, Majoo, ESB)
  • Phased rollout: API sync → shadow-mode validation → active recommendations
  • Supplier B2B integration (year-2 roadmap)
  • Regional expansion in Southeast Asia

Not yet validated Unvalidated

  • Forecast accuracy (calibration planned over a client’s first 4–6 weeks)
  • Measured food-waste reduction
  • Market-size, pricing, and revenue projections (indicative only)