MySmartCoPilot
Ask Passes Your pass 7 days left 1 day left Last day

Your country

Tools that support it use your country for local currency, number formats, units and paper size. Your choice is saved only in this browser.

Go Module 4

Arrays, slices, maps and strings interview questions

Interview questions on arrays, slices, maps and strings: warm-ups, then what each harder question tests and how a good answer starts.

23 questions: 6 warm-ups and 17 with model answers · Levels: Fresher to senior

Warm-up questions

Short questions with short answers, each answered in full.

  1. Fresher

    What happens when you assign one array variable to another in Go?

    Answered in the lesson Arrays are values: length, copies, comparison

  2. Fresher

    What does a slice value hold?

    Answered in the lesson Slices: length, capacity, append and aliasing

  3. Fresher

    Why does slices.Sort put "Zoya" before "arjun"?

    Answered in the lesson The slices package: sort, search and compare

  4. Fresher

    How do you tell a missing key from a key that holds the zero value?

    Answered in the lesson Maps: lookups, sets, iteration and gotchas

  5. Fresher

    Why is len("héllo") 6 in Go?

    Answered in the lesson Strings, bytes and runes: UTF-8 done right

  6. Fresher

    What does strconv.Atoi return for "42x"?

    Answered in the lesson Numbers as text: strconv, bases and grouping

Questions with model answers

What each question tests, the first step of a good answer and the follow-ups to expect.

  1. Fresher to mid

    A function func applyGrace(marks [5]int) adds 5 to every mark below 40, but the report printed after the call still shows the old marks. Explain what happened, then give two fixes and say when you would choose each.

    What it tests
    Knowing that an array parameter is a copy, fixing it with a returned array or a pointer, and choosing between them and a slice by size and intent.
    First step
    Say what the call passes: marks is an array, so applyGrace receives its own copy of all five marks, changes that copy and drops it when it returns. Then choose how the change reaches the caller: return the new array and assign it, or take a pointer, *[5]int, and write through it.
    Follow-ups

    From the lesson Arrays are values: length, copies, comparison

    Opening the model answers with your pass…

    Model answers open with any MySmartCoPilot pass.

    • Pro pass · 1 week₹79$2.99
    • 30 days₹179$5.99
    • 1 year₹1,699$45

    One payment; a pass never renews. The price shown is the total you pay: no fees or charges are added at checkout.

    Choose a passI have a licence key

    It also opens every other in-depth part in Learn and every Pro tool. Sales are final, except where the law gives you a right to withdraw, for example in the EU and the UK (refund policy).

    In the answer:

    • Model answer
    • Follow-up answers
    • Common mistakes
    • Rubric
    • Variants
    • about 3 minutes

  2. Mid to senior

    A service must find duplicate uploads among 50,000 files by their SHA-256 hashes. Why can the [32]byte that sha256.Sum256 returns be a map key when a []byte cannot, and how do you design the lookup?

    What it tests
    Knowing which types are comparable and why map keys need them, using fixed-size arrays as keys, hashing large files without loading them, and estimating the memory of the index.
    First step
    Start from the rule: a map lookup compares keys with ==, so the key type must be comparable. An array of bytes is comparable element by element, and a slice is not comparable at all. Keep the [32]byte as it is and use map[[32]byte]string, from each hash to the first file that had it.
    Follow-ups

    From the lesson Arrays are values: length, copies, comparison

    In the answer:

    • Model answer
    • Follow-up answers
    • Common mistakes
    • Rubric
    • Variants
    • about 3 minutes

  3. Fresher to mid

    A function builds two queries from one base slice of filters: q1 := append(base, "status=open") and q2 := append(base, "status=closed"). In production both queries end with status=closed, but the unit test, which passes a slice literal as base, passes. Explain both results and fix the function.

    What it tests
    Knowing that append re-uses spare capacity, why the result depends on how base was built, and fixing it with a full slice expression, slices.Clip or a copy.
    First step
    Look at the capacity of base, not at its elements. A literal such as []string{"a", "b"} has no spare room, so each append copies and the test passes. The production base was built by appends and has room, so both appends write the same free element, and the second write replaces the first.
    Follow-ups

    From the lesson Slices: length, capacity, append and aliasing

    In the answer:

    • Model answer
    • Follow-up answers
    • Common mistakes
    • Rubric
    • Variants
    • about 3 minutes

  4. Mid to senior

    A service reads 64 MiB log files whole, keeps the first 32 bytes of each as its header in a map and drops the rest. After 100 files the heap holds over 6 GiB. Explain why, say how you would confirm it, and fix it.

    What it tests
    Knowing that a slice keeps its whole backing array alive, confirming it with a heap measurement or profile, and copying out, or reading, only what is kept.
    First step
    Follow what the map holds: each header is data[:32], a slice whose pointer points into the 64 MiB array of its file. While the map holds the slice, the collector must keep that whole array, so 100 headers keep 100 × 64 MiB, 6,400 MiB in all.
    Follow-ups

    From the lesson Slices: length, capacity, append and aliasing

    In the answer:

    • Model answer
    • Follow-up answers
    • Common mistakes
    • Rubric
    • Variants
    • about 3 minutes

  5. Fresher to mid

    RemoveAt(s []int, i int) []int returns append(s[:i], s[i+1:]...). A caller removes index 2 from [1 2 3 4 5], prints its own slice and sees [1 2 4 5 5]. Explain the output, then design the function's API.

    What it tests
    Knowing how an in-place removal shifts elements inside the caller's array, reading the contract of slices.Delete, and choosing between an in-place and a copying API.
    First step
    s[:i] keeps the caller's array and capacity, so the append writes 4 and 5 over the caller's elements 2 and 3 and returns a header of length 4. The caller's own header still has length 5, so it shows the shifted elements followed by the old last element.
    Follow-ups

    From the lesson Slices: length, capacity, append and aliasing

    In the answer:

    • Model answer
    • Follow-up answers
    • Common mistakes
    • Rubric
    • Variants
    • about 3 minutes

  6. Fresher to mid

    A code review finds slices.SortFunc(accounts, func(a, b Account) int { return int(a.Balance - b.Balance) }) with int64 balances, where math.MinInt64 marks an account with no balance yet. The tests pass. What can go wrong, and what do you write instead?

    What it tests
    Knowing that a comparison must give the right sign for every pair, that integer subtraction wraps on overflow, and using cmp.Compare and cmp.Or.
    First step
    Check the sign of a.Balance - b.Balance for extreme values. For 500 and math.MinInt64 the true difference is about 9.2 billion billion, past the largest int64, so it wraps to a negative number and says 500 sorts before the marker. The sort trusts the sign, so the result is out of order, with no error.
    Follow-ups

    From the lesson The slices package: sort, search and compare

    In the answer:

    • Model answer
    • Follow-up answers
    • Common mistakes
    • Rubric
    • Variants
    • about 2 minutes

  7. Mid

    A transactions page sorts its rows by date with slices.SortFunc. Users report that rows with the same date change places when they refresh, but the tests, which use ten rows, always pass. Explain both observations and fix it.

    What it tests
    Knowing that SortFunc is not stable, why a small test can still look stable, and fixing the order with SortStableFunc or a final key that never ties.
    First step
    Name the guarantee: SortFunc's documentation does not promise to keep equal elements in their order. Rows with one date tie, so their order depends on the sort's moves, and on the order the rows arrived in from the database, which can differ between requests. Then ask what the sort does differently with ten rows than with hundreds.
    Follow-ups

    From the lesson The slices package: sort, search and compare

    In the answer:

    • Model answer
    • Follow-up answers
    • Common mistakes
    • Rubric
    • Variants
    • about 3 minutes

  8. Mid to senior

    A service checks each request against one million blocked ids kept in a sorted slice with slices.BinarySearch. A new feature appends ids to the end of the slice as they are reported, and some blocked ids now get through. Why, and how do you keep the check correct and fast?

    What it tests
    Knowing that binary search requires sorted input and fails silently, the cost of inserting into a sorted slice, and choosing between sorted inserts, batch sorting and a map.
    First step
    Recall what the search assumes: the slice is sorted, so each look throws away half of it. An appended id that is smaller than its neighbours sits in a part the search has already thrown away, so it reports not found without any error. The check is now wrong for exactly the newest ids.
    Follow-ups

    From the lesson The slices package: sort, search and compare

    In the answer:

    • Model answer
    • Follow-up answers
    • Common mistakes
    • Rubric
    • Variants
    • about 3 minutes

  9. Mid to senior

    An HTTP service keeps sessions in a package-level map[string]Session that every handler reads and writes. Under load it dies with fatal error: concurrent map writes, and the recover in its middleware never runs. Explain why, then fix it and say how you would catch it in tests.

    What it tests
    Knowing that maps are unsafe under concurrent writes, what the runtime's check does to a program, choosing between a mutex and sync.Map, and testing with the race detector.
    First step
    Handlers run in their own goroutines, so two requests can write the map at once, which Go does not allow, and the runtime's check stopped the program. Read the first word of the message, fatal, and ask what a fatal error does to deferred calls, which is where the middleware's recover lives.
    Follow-ups

    From the lesson Maps: lookups, sets, iteration and gotchas

    In the answer:

    • Model answer
    • Follow-up answers
    • Common mistakes
    • Rubric
    • Variants
    • about 3 minutes

  10. Mid to senior

    A cache map held 10 million entries at a traffic peak. A nightly purge deletes all but 50,000 of them, yet the process's heap stays near its peak for days. Explain what you see, and fix it without restarting the process.

    What it tests
    Knowing that a Go map keeps its storage after deletes and clear, measuring it, and rebuilding or bounding the map safely while it is in use.
    First step
    Measure before guessing: after a garbage collection the heap stays at the peak, so something still owns the memory. Look at the map itself: deleting an entry frees its slot for a later insert, but nothing says the map gives back the storage it grew to at the peak.
    Follow-ups

    From the lesson Maps: lookups, sets, iteration and gotchas

    In the answer:

    • Model answer
    • Follow-up answers
    • Common mistakes
    • Rubric
    • Variants
    • about 3 minutes

  11. Fresher to mid

    byUser is a map[string]Stats, and byUser[id].Visits++ does not compile. Explain the error, give two ways to make the change, and say when each fits.

    What it tests
    Knowing that map elements are not addressable, writing a copy back or storing pointers, and the trade-offs of each, including range loops that change copies.
    First step
    Read the message: cannot assign to struct field byUser[id].Visits in map. Ask what the left side of an assignment may be, and whether a map element is one of those things; both ways to make the change follow from the answer.
    Follow-ups

    From the lesson Maps: lookups, sets, iteration and gotchas

    In the answer:

    • Model answer
    • Follow-up answers
    • Common mistakes
    • Rubric
    • Variants
    • about 3 minutes

  12. Mid

    A notification service shortens each message to fit a 160-byte field with msg[:160]. Messages in Hindi sometimes end in a replacement character on phones. Explain what happens, fix it, and say what can still go wrong with emoji.

    What it tests
    Knowing that slicing a string cuts bytes, that UTF-8 characters take several bytes, cutting at a rune boundary, and the limits of rune-level cutting for grapheme clusters.
    First step
    Count bytes, not letters: each Devanagari code point takes three bytes in UTF-8, so a cut at byte 160 can land inside one, leaving one or two of its bytes at the end. Those bytes are invalid UTF-8, and the phone shows the replacement character in their place.
    Follow-ups

    From the lesson Strings, bytes and runes: UTF-8 done right

    In the answer:

    • Model answer
    • Follow-up answers
    • Common mistakes
    • Rubric
    • Variants
    • about 3 minutes

  13. Fresher to mid

    A report builder adds 50,000 lines of about 40 bytes to a string with report += line, and the job is slow and uses far more memory than the 2 MB report. Explain the cost with numbers and rewrite it.

    What it tests
    Knowing that strings are immutable so += copies the text so far, estimating quadratic copying, and using strings.Builder, Grow and strings.Join.
    First step
    Follow one +=: a string cannot change, so it allocates a new string and copies the whole report so far plus the line. The k-th line copies about k × 40 bytes, so 50,000 lines copy about 40 × 50,000² / 2 bytes, some 50 GB, for a report of 2 MB.
    Follow-ups

    From the lesson Strings, bytes and runes: UTF-8 done right

    In the answer:

    • Model answer
    • Follow-up answers
    • Common mistakes
    • Rubric
    • Variants
    • about 3 minutes

  14. Mid to senior

    A sign-up form must limit usernames to 20 characters. The Go check len(name) > 20 rejects many names written in Devanagari that look short. What should "characters" mean here, and how do you implement and test the limit?

    What it tests
    Knowing the difference between bytes, runes and grapheme clusters, choosing the unit a rule means, validating UTF-8 and normalising input, and testing with scripts beyond ASCII.
    First step
    Name the three units: len counts bytes, three per Devanagari code point, so a 7-letter name can pass 20 bytes; utf8.RuneCountInString counts code points; what a user sees as a character is a grapheme cluster, which can be several runes. Decide which one the rule means, then measure that.
    Follow-ups

    From the lesson Strings, bytes and runes: UTF-8 done right

    In the answer:

    • Model answer
    • Follow-up answers
    • Common mistakes
    • Rubric
    • Variants
    • about 3 minutes

  15. Fresher to mid

    A CSV import reads quantities with strconv.ParseInt(field, 10, 8) to save memory, logs any error and stores the returned value. A row with 300 is stored as 127. Explain what happened and fix the import.

    What it tests
    Knowing that ParseInt returns the limit of the bit size together with ErrRange, never using a value after an error, and choosing a bit size from the data.
    First step
    Read what ParseInt promises on a range error: the returned value is the largest magnitude of the bit size and sign, 127 for 8 bits, together with ErrRange. The import logged the error but stored the value anyway, so every quantity above 127 became 127, and every one below -128 became -128.
    Follow-ups

    From the lesson Numbers as text: strconv, bases and grouping

    In the answer:

    • Model answer
    • Follow-up answers
    • Common mistakes
    • Rubric
    • Variants
    • about 3 minutes

  16. Mid

    An API answers bad query parameters with the error text from strconv, such as strconv.ParseInt: parsing "12a": invalid syntax. What is wrong with that, and how do you turn a *strconv.NumError into good messages?

    What it tests
    Knowing NumError's fields and errors.Is with ErrSyntax and ErrRange, writing messages that name the field and the rule, and keeping internal names and raw input out of responses.
    First step
    Read it as the client does: it names a Go function the client never called, and says neither which parameter was wrong nor what is allowed. The error holds what a better message needs: errors.Is with strconv.ErrSyntax or strconv.ErrRange gives the kind of mistake, and the code that parsed knows the field and its rule.
    Follow-ups

    From the lesson Numbers as text: strconv, bases and grouping

    In the answer:

    • Model answer
    • Follow-up answers
    • Common mistakes
    • Rubric
    • Variants
    • about 3 minutes

  17. Fresher to mid

    Write FormatAmount(paise int64) string, which prints amounts with Indian digit grouping, such as 12,34,567.89. Which edge cases must its tests cover, and why does each one exist?

    What it tests
    Knowing integer division and remainder for minor units, the grouping rule, the sign of remainders, and the int64 edge at math.MinInt64.
    First step
    Keep the amount an integer: the rupees are the amount divided by 100, and the paise are the remainder, printed with two digits. Group the rupee digits from the right, three first and then pairs. Then test the obvious version at the edges of int64 and with negative amounts before you trust it.
    Follow-ups

    From the lesson Numbers as text: strconv, bases and grouping

    In the answer:

    • Model answer
    • Follow-up answers
    • Common mistakes
    • Rubric
    • Variants
    • about 2 minutes

All Go interview questions · The Go track