This page maps real code, data pipelines, and infrastructure to the exact skills and roles demonstrated by FleetFlow.
Live demo runs on free-tier infrastructure
Supabase + Upstash + Railway (fallback-safe, limited workloads)
Full-power stack is local-first
Run locally to unlock Kafka, Flink, Airflow, PostGIS, TimescaleDB, Redis, MongoDB, OSRM.
Producers emit orders, GPS, and route events into Kafka topics for real-time flow.
Operational data is stored in Postgres/PostGIS with time-series in TimescaleDB.
Airflow DAGs orchestrate dbt models and scheduled optimization workloads.
OR-Tools and custom heuristics compute multi-objective routes and assignments.
Forecasting ensemble blends Prophet, LSTM, and XGBoost for demand prediction.
FastAPI APIs, WebSockets, and monitoring dashboards expose real-time state.
Senior-grade foundations
Applied forecasting + EDA
Experimentation + serving scaffolding
Production-style API core
Type-safe dashboard + real-time UI
Multi-store + infra blueprints
The live demo is optimized for free-tier deployments. Clone the repo and run locally to unlock the full stack and advanced workloads.