Node.js

Imagine a CRM system: leave requests are approved, but absences are not marked in the calendar, causing incorrect salary calculations. The problem is floating. How would you start solving this problem?

How exactly do we accept and save an event — how should the reception layer be organized, and can we trust the event?

Describe my most interesting case

Part of the work involves analyzing and refining existing code, sometimes generated by AI, sometimes without documentation. How do you feel about this format?

Imagine a system that receives events from an external system (payments, order statuses) via HTTP webhook or queue. It needs to save events, update the state, and do this reliably in production. How would you design the overall flow scheme?

What should I answer to the question "what are you looking for in the context of the vacancy"?

What if a Success/Paid/Done event arrives first, and then an older Pending event? How to resolve this?

How to distinguish a temporary error from one that cannot be recovered?

Как обеспечивал exactly-once доставку, или это только про консистентность?

An external system sent the same event multiple times. How to solve this problem?

Suppose you diagnosed a problem. How can you make sure you don't have to troubleshoot manually next time?

If everything is checked and the discrepancy still exists — where else could the error be, and could the provider itself be at fault (for example, if their API is not delivering correctly)?

Tell me about a project or something you did that you are proud of, and why?

Suppose everything you checked is correct. What then?

The system is already in production. How to understand that it is working, and how to investigate a specific problem?

The external provider sends webhooks, we record in the database. After a month, the product comes: they have 1000 records, we have 970. How will you investigate?

Describe my experience that will help me in the interview for the position

The event is valid, not a duplicate, the order is clear, but the service crashes during processing. How to ensure reliability?

What tables and entities are needed in the database to ensure the system is maintainable and auditable?

What to show in the UI if the event is saved but not applied?