Go for infrastructure and high-throughput services, where concurrency idioms carry most of the interview weight.
8 published across 4 topics.
Go fundamentals60 short answers on one page, for revising rather than studying.Slices and maps internals, interfaces and embedding, error handling conventions, and generics.
An interface value carries a type alongside a pointer, so an interface holding a nil pointer of a concrete type is itself not nil. Returning a typed nil pointer into an error return produces a value that is non-nil to every caller while being nil to the function that produced it.
append writes into the existing backing array when spare capacity exists and allocates a fresh array only when it does not, so whether two slices still alias each other after an append depends on capacity rather than on anything visible at the call site.
The scheduler, channel semantics, select, context cancellation, and detecting goroutine leaks.
A goroutine leaks when it blocks forever on an operation nobody will complete: a send with no receiver, a receive on a channel nobody closes, or a worker abandoned after its consumer returned early. Fix it by giving every blocking operation a context escape hatch, and find it with pprof goroutine profiles.
The sender closes, never the receiver, and with several senders no individual one may close because a send on a closed channel panics. The answer is a WaitGroup over the producers with a separate goroutine closing once they have all finished, which is the only place the fact 'nobody will send again' actually exists.
Escape analysis, allocation reduction, GC behaviour, and pprof-driven profiling.
Escape analysis moves a value to the heap when the compiler cannot prove its lifetime ends with the frame that created it, and go build -gcflags=-m prints every decision. Which of those allocations matter is a measurement question, answered with a benchmark, an alloc_space profile and benchstat.
Take a heap profile and compare two taken minutes apart, because the shape of the growth tells you whether it is a leak, a cache with no bound, or simply the collector letting the heap grow. The two real knobs are GOGC and GOMEMLIMIT, and reducing allocation matters more in Go than in a generational runtime.
Standard-library HTTP, project structure, dependency management, and testing patterns.
Every incoming request carries a context that the server cancels when the client disconnects or the handler returns. You thread it as the first parameter of every call, narrow it per dependency with context.WithTimeout, and back it with server-level timeouts and a graceful Shutdown.
Wire dependencies explicitly through constructors from one place in main, define narrow interfaces where they are consumed rather than implemented, and keep handlers thin enough that the logic beneath is testable without a server. With no framework, these are decisions rather than defaults.