DevFlow OS is a premium, multi-agent AI-powered Innovation Operating System designed to guide student teams, startups, and innovators from initial concept discovery to presentation-ready Samsung Solve for Tomorrow submissions.
Built on a Next.js 16 monorepo and styled with a sleek, dark Samsung-inspired glassmorphism design language, it bridges the gap between chaotic brainstorming and structured, compliance-ready project frameworks.
DevFlow OS models the double-diamond Design Thinking methodology combined with active AI validation gates.
flowchart LR
A[1. Problem Discovery] --> B[2. Empathise Stage]
B --> C[3. Define Stage]
C --> D[4. Ideate & Model]
D --> E[5. Prototype CAD/Code]
E --> F[6. User Validation]
F --> G[7. AI Council Review]
G --> H[8. Generate Report & Widescreen Deck]
- Innovation Workspace: A premium flight deck hosting dynamic project progress indices, health rings, and a double-diamond design thinking wizard.
- Problem Discovery Panel: Tools for root cause analysis including the 5 Whys drill-down, stakeholder maps, chronological research logs, and visual Fishbone skeletons.
- AI Innovation Mentor (NOVA): Continuous contextual coaching that flags product blind spots, estimates feasibility, assesses risks, and maps SDG alignment.
- Spatial Innovation Galaxy Map: A beautiful 3D particle orbit representation where projects revolve around the AI Core. Orbit distance represents development stage, and connection lines trace shared SDG paths.
- Impact Intelligence Matrix: Formulates target SDG weights, outputs predictive beneficiary growth curves, and highlights inclusivity hazards.
- NOVA Expert Council: A simulated panel of 8 specialized AI agents (Tech Lead, SDG Ethicist, Accessibility Expert, and others) performing automated evaluations to deliver a unified Readiness Score.
- One-Click Slide Deck & Reports: Instantly render high-contrast pitch decks and compile print-ready PDF reports containing empathy maps, SDG matrices, and AI council reviews.
DevFlow is organized as a decoupled monorepo leveraging npm workspaces and Turborepo:
graph TD
User([User Client]) -->|HTTP Port 3000| FE[Frontend Next.js App]
FE -->|REST API Port 8000| BE[Backend FastAPI Gateway]
BE -->|SQL Port 5432| DB[(PostgreSQL Database)]
BE -->|TCP Port 6379| RD[(Redis Broker / Cache)]
subgraph local_network [Local Docker Bridge Net]
BE
DB
RD
end
- Frontend Client:
http://localhost:3000 - Backend API Gateway:
http://localhost:8000 - Interactive Swagger Docs:
http://localhost:8000/docs - PostgreSQL Database:
localhost:5432 - Redis Cache & Broker:
localhost:6379
Make sure your local host has the following runtimes installed:
- Node.js (v20+ / npm 10+)
- Python (v3.13+)
- Docker & Docker Compose (Recommended for full services)
Run the automated installer script from the root folder:
make installUnder the hood, this script (scripts/setup_dev.py) automatically generates environment config files, configures python virtual environments, installs npm packages, and sets up Git pre-commit format hooks.
Compile and launch the containerized stack:
# Start the services
make start
# Verify container health checks (pg_isready, api health endpoints)
docker compose psTo shut down and wipe volumes:
make stopIf you prefer running services natively:
- Start your local PostgreSQL and Redis servers.
- Run the concurrent developer pipeline:
make devThis runs the Next.js client on port 3000 and the FastAPI backend on port 8000 with hot-reload watch processes enabled.
Maintain codebase stability by running verification routines:
- Format Code:
make format(runs Prettier for tsx/css, Black/Ruff for python) - Static Linting:
make lint(runs ESLint and Mypy static check compilers) - Unit Tests:
make test(runs Vitest for client, Pytest for backend)
βββ apps
β βββ api # FastAPI Python backend service
β β βββ app # Core logic (endpoints, middleware, schemas, models)
β β βββ tests # Backend testing suite
β βββ web # Next.js App Router frontend application
β βββ src # Frontend assets, hooks, styling, and components
β βββ public # Static images and client data assets
βββ packages # Shared configurations
β βββ config # Shared ESLint, Prettier, and Typescript configs
β βββ types # Global typings and models
β βββ ui # Shared layout frameworks
βββ docs # Production blueprints, architecture, and diagrams
βββ scripts # Automated verification and installation utilities
βββ Makefile # Command shortcut orchestration
βββ turbo.json # Monorepo task cache configurations
Refer to our deep-dive architecture folders:
- System Architecture Guide
- Front & Backend Design Specifications
- REST Endpoint Catalog
- Git Rules & Contributing Guidelines
- Production Deployment Protocols
- Security & Trusted Host Configurations
- Product Milestone Changelog
- Project Future Roadmap
Contributions are welcome! Please review CONTRIBUTING.md for details on code style requirements, branch patterns, and pull request validations.
This project is licensed under the MIT License - see the LICENSE file for details.
For Samsung Solve for Tomorrow queries, demo day schedules, or general support, please reach out to the project maintainers at support@devflow.io.