In short
HackerRank holds 4.5 out of 5 on G2 from 576 reviews, and teams praise its integrity flags and ease of use. They criticise the question library, the IDE and AI quality, so the questions to settle are library fit and whether you can check the AI output. Tracy AI Interviewer drafts questions from your job description and keeps a recording and transcript behind every result. Your team decides.
On G2, HackerRank holds 4.5 out of 5 from 576 reviews. This page reads the first 10 reviews G2 listed on 14 November 2026, dated 28 June to 12 November 2026. Seven come from business users that G2 counts in its scores, and three are flagged by G2 as outside them. Reviewers are engineering managers, talent acquisition staff, a senior architect and engineers. The counts below cover all 10; where leaving out the flagged three changes a number, the text says so. HackerRank's plans and prices are covered separately.
The reviews are about tests, flags and the pad. How each applicant explains their own work is the part Tracy takes.
What do buyers like about HackerRank?
Integrity flags were the most common praise: 6 of 10 reviews mention them. An engineering manager wrote that HackerRank "can flag candidates who may be trying to cheat" (G2, 15 Sep 2026), and a senior architect liked the "advanced Proctor feature" (G2, 28 Jun 2026).
Ease of use came next, in 5 reviews. A reviewer called the platform "simple for candidates to use" (G2, 9 Oct 2026), and another said it "looks and feels like Visual Studio Code or Cursor" (G2, 11 Nov 2026).
Four reviews like building their own questions and assessments, and four like the live code experience, the IDE and the replay. One reviewer wrote that "the live interview/CodePair experience" lets them watch a candidate think in real time (G2, 8 Sep 2026). A talent acquisition specialist said the cost is "worth the hours we save" in engineering time (G2, 28 Oct 2026).
That praise is about tests, the pad and the flags. A conversation that follows each applicant's own work is a different job.

Reviewers praise integrity flags and ease of use most, and criticise the question library and the IDE and interface equally.
What do buyers complain about?
Three themes reached three reviews or more.
The question library: 4 reviews. One reviewer found that "the questions start getting a bit similar" over time (G2, 21 Sep 2026). A senior architect said out-of-the-box questions are "either too hard" or "too easy", so the team writes its own (G2, 28 Jun 2026). Another found the library "can feel generic for more specialized roles" (G2, 8 Sep 2026). Counting only the 7 reviews G2 includes in its scores, the library theme is in 3.
The IDE and interface: 4 reviews. Comments ranged from Excalidraw feeling "a bit more polished" for whiteboarding to the CodePair interface feeling "clunky" and the IDE being "buggy/unstable sometime" for some candidates. Counting only the 7 scored reviews, 3 raise it.
AI quality: 3 reviews. A senior architect called the latest AI efficiency score "unreliable" (G2, 28 Jun 2026), another was "not convinced the AI is that good" (G2, 8 Nov 2026), and a third said people are confused about which AI model to use (G2, 11 Nov 2026). Counting only the 7 scored reviews, 2 raise it.
Price came up in 2 reviews, link expiry management in 1 and ATS sync in 1, too few to call a theme.
What do reviewers say about price, ATS sync and link expiry?
Price came up in 2 reviews. One reviewer, a senior architect, wrote "A bit more expensive" (G2, 28 Jun 2026). Another, flagged by G2 as outside its scores, said "upgrading to get basic proctoring features costs too much for smaller teams" (G2, 9 Oct 2026). A third says the cost is worth the engineering hours saved, as quoted above.
The same flagged reviewer said "the system sometimes fails to update our hiring software automatically", so scores are checked by hand (G2, 9 Oct 2026). A staffing reviewer wished there were "ways to extend tests' expiration dates more efficiently than having us resend the link" (G2, 5 Nov 2026). Each is a single review, so treat them as questions for a rep, not as a pattern.
What each review theme means for a buyer
Counts from the first 10 G2 reviews listed on 14 Nov 2026.
| Theme | Reviews | What to do with it |
|---|---|---|
| Question library | 4 of 10 | Check the library against your roles; plan to write your own questions |
| IDE and interface | 4 of 10 | Run a trial session as a candidate before you commit |
| AI quality | 3 of 10 | Ask how the AI output is produced and who reviews it |
| Integrity flags | 6 of 10 praise it | A strength; a person still reviews every flag |
Who does HackerRank suit?
HackerRank suits teams that want one vendor for timed assessments, live pair programming and proctoring, and that have the engineers to write questions for specialised roles. Its pricing page lists SOC 2, ISO 27001 and GDPR on all plans, 22+ ATS integrations on Enterprise, and a Chakra AI interviewer on every plan.
It suits a team less well if your roles are narrow and the library feels generic, or if you want to see how a person reasons about their own work rather than how they solve a standard problem. If you cannot yet check AI output, ask how it is produced before you rely on it.
What does HackerRank cost?
HackerRank's pricing page, read 15 Nov 2026, lists Starter at $79 a month billed annually for 60 candidate attempts a year, Pro at $419 a month ($5,028 billed annually) for 360 attempts with ATS and calendar integrations, and Enterprise on request. HackerRank's FAQ says a candidate completing an attempt through Screen, Interview or Chakra counts against the attempt credit, so a Chakra session uses one attempt. Usage beyond the plan is charged at the end of the billing cycle or after 25 extra attempts, whichever comes first. Extra attempts are $20 each, or $15 each when pre-purchased on an annual Pro plan. The full breakdown, with cost at three volumes, is in HackerRank pricing.
What should you ask HackerRank before you buy?
Before you buy, ask HackerRank three things. Which questions in the library cover your roles, and how many would your engineers still write? What can you do when a candidate misses the window, and how do you extend a test's expiry without resending the link? How is the AI efficiency score produced, and does a person review it? Ask for a trial session as a candidate too, since the interface and IDE are among the two most common criticisms.
Which review theme a recording and transcript answer
The problem is AI output that you cannot check. 3 of the 10 reviews raise AI quality, and a senior architect called the latest AI efficiency score "unreliable" (G2, 28 Jun 2026). Counting only the 7 scored reviews, 2 raise it.
A score or summary can only be checked against the words behind it. So before relying on any AI judgment, ask to open the recording and the transcript at the point it refers to. In a Tracy result each written assessment sits beside both, and a person reads them and decides.
How does Tracy answer the concerns reviewers raise?
In the sample, the most frequent concerns are a library that feels generic and the IDE and interface, 4 reviews each; AI output that is hard to check came third, with 3. Tracy AI Interviewer drafts questions from the job description and lets you edit them or bring your own, so the first question belongs to the role. It can read the CV when the CV check is on, asks follow-ups in the candidate's own field and puts a coding task inside the interview. Coding interview follow-up questions shows what those follow-ups sound like.
Behind every Tracy result sit a recording of both sides, a timestamped transcript and a written assessment per question with its reasoning, so a reviewer can check the AI's work against the words. Tracy reports what was said and a person decides.
Since price came up in reviews: Tracy costs $1 per credit with no seats, and the first 3 results are free. The cost at several volumes is in AI interview cost. HackerRank is the stronger pick for live pair programming, proctoring depth and a large question library. For a side-by-side, read Tracy vs HackerRank, or see how technical interviews run in Tracy.
What Tracy adds to technical interviews
01
Reach every applicant
Each applicant can start on click, at any hour, with no scheduling and no recruiter on the call. Engineers are not booked for the first round.
02
Ask the next question
Tracy follows the answer in the applicant's own field and waits. A coding task sits inside the interview, and Tracy asks about the decisions behind it.
03
Keep the evidence
Each result comes with a recording, a timestamped transcript and a written assessment per question with its reasoning. A reviewer can check the AI's work against the words.

From job post to a result you can check against the words.

Tracy follows a technical answer with a question about the decision behind it.
Common questions
Is HackerRank good, according to reviews?
What do HackerRank reviews say about the question library?
What do HackerRank reviews say about price?
Are there HackerRank employee or Glassdoor reviews on this page?
What is a HackerRank alternative for technical interviews?
Read one result next to HackerRank's
Create one interview for a role you screen now and compare its report with your current tool. The first 3 results are free.
Book a demo (opens in a new tab) Try it free (opens in a new tab)
Your first 3 results are free. No card needed.
Sources
- G2, HackerRank Developer Skills Platform reviews, retrieved 14 Nov 2026.
- HackerRank, pricing, retrieved 15 Nov 2026.
- Tracy AI Interviewer, Pricing, retrieved 14 Nov 2026.
- Tracy AI Interviewer, Connect, retrieved 14 Nov 2026.



