Node.js Accounting Platform Overhaul

Our cloud-based accounting platform is built entirely with a JavaScript/Node.js back-end on top of MongoDB. During closed beta everything behaved, but the moment we opened the doors to a larger audience the application locks up under load—no gradual slowdown, just a complete system freeze that forces a hard restart.

I need a seasoned engineer who can dedicate roughly 5–6 hours every day for at least the next three months to hunt down the root causes and redesign the platform so it scales smoothly. You’ll collaborate directly with me and two in-house developers; we work on GitHub, communicate over Slack, and deploy through a basic CI pipeline that will probably need tightening as part of this project.

Key objectives
1. Diagnose and eliminate the freeze: trace blocking operations, long-running Mongo queries, memory leaks, or event-loop starvation.
2. Re-architect for scale: propose and implement a more resilient design—whether that means service decomposition, queueing, load balancing, horizontal sharding, or smarter caching.
3. Harden the codebase: add automated tests, performance monitoring, logging, and graceful-degradation fallbacks.

Acceptance criteria
• Platform remains responsive under sustained load equivalent to our projected peak (we’ll define exact numbers together).
• No unhandled crashes or freezes for seven consecutive days in staging and production.
• Clear documentation of all architectural changes and runbooks for future maintenance.

If tackling deep Node.js performance puzzles and large-scale MongoDB optimisations excites you, let’s talk.