Pragmatic Engineering: Pivoting from Stack Fantasy to Product Constraints

Privacy-first mind mapping app. Part 1: Constraints Before Tech

Kornel Maraz initially attempted to build MindMapVault using a philosophically pure, end-to-end Rust stack. This approach was abandoned after experimental components like RustFS failed to maintain stability at 200 concurrent users.

Why This Matters

The transition from a ‘technology demo’ to a practical product highlights the danger of stack fantasy, where technology choices are treated as the product itself. Engineers must prioritize constraints—such as iteration speed, memory footprint, and component availability—to avoid the trap of reinventing mature tooling while building early-stage SaaS applications.

Key Insights

• Scalability failure of RustFS: The object storage tool proved unstable under load, failing at 200 concurrent users during the 2026 alpha phase.
• Memory footprint constraints: MongoDB and Python frameworks were rejected to minimize operational costs and memory usage for a long-running SaaS.
• Frontend iteration speed: React and TypeScript were selected over Rust web UI tools to leverage mature mind-mapping canvas components.
• Backend performance: Rust with Tokio and Axum was retained as a practical middle ground to ensure high responsiveness and low overhead.
• Desktop integration: Tauri was utilized to bridge the web frontend with a secure, lightweight desktop application while maintaining Rust synergy.

Practical Applications

• Use case: Utilizing MinIO for S3-compatible object storage to ensure boring but reliable persistence. Pitfall: Using experimental storage like RustFS in early production leads to instability under concurrent load.
• Use case: Implementing PostgreSQL as the storage baseline for proven reliability. Pitfall: Managing multiple database experiments like SurrealDB during early-stage development creates unnecessary documentation and skill hurdles.
• Use case: Deploying Tauri for cross-platform desktop apps to reuse web code efficiently. Pitfall: Building complex UI components from scratch in immature ecosystems significantly delays the time-to-market.

References:

• https://dev.to/kornel_maraz_5e66a3e4e27d/privacy-first-mind-mapping-app-part-1-constraints-before-tech-50p3