FlowFrame

While learning system design, I often found static diagrams difficult to reason about.

A diagram can show components like Clients, Load Balancers, API Gateways, Redis, and Databases—but it doesn't show what actually happens when a request moves through the system.

FlowFrame was built to make distributed systems easier to understand through interactive simulations.

Current simulations include:

• Load Balancing
• Cache-Aside Pattern (Redis + PostgreSQL)
• API Gateway Routing
• Valet Key Pattern

Features:

• Interactive request flow visualization

• Frame-by-frame playback

• Node state inspection

• Simulation logs with timestamps

• Sandbox environment for experimentation

FlowFrame already supports a variety of request and routing behaviors, including endpoint validation, routing outcomes, and component-specific failure scenarios across simulations.

Users can inspect logs, step through frames manually, and observe how requests propagate through the system over time.

One area I'm especially excited about is expanding failure simulation beyond request-level failures into distributed system behaviors such as:

• Redis latency or downtime

• Database connection pool exhaustion

• Retry storms

• Queue backpressure

• Cache TTL expiration

• Service degradation and failover scenarios

The goal is to help developers understand not only the happy path, but also how systems behave under failure conditions—making the platform useful for both learning and interview preparation.

This is currently an early MVP, and I'm actively looking for feedback from developers, backend engineers, and system design learners.

I'd love feedback on:

• What distributed system pattern should be added next?

• Which failure scenarios would be most valuable?

• Would an interactive approach help you learn system design more effectively?