CodeSignal is a skills platform for technical hiring. Candidates write and run code in a real IDE, take agentic assessments with tools like Claude and Cursor, and sit AI or live interviews, with proctoring and ID checks on every plan. TunnelQuiz is a narrower tool: proctored multiple-choice and written tests for any role, priced per candidate, with screen and browser checks from the free plan up. If you hire engineers at volume, CodeSignal is the stronger product. If most of your screening is aptitude, language and domain knowledge, TunnelQuiz will cost you far less per candidate.
CodeSignal's strength is evaluating engineers: code that runs in a built-in IDE, 2,000 library questions on its entry plan, agentic assessments with Claude, Cursor and Codex, and live technical interviews. TunnelQuiz is built around multiple-choice and written tests. It does take coding questions in 19 languages on Pro, but an AI grades them and the code is never run. If your process is code-heavy, CodeSignal is purpose-built for it.
CodeSignal's Build plan is US$79 a month billed annually for 60 credits a year, or $99 a month for 5 credits a month, and each extra credit costs $20. TunnelQuiz's Essential plan includes 150 credits every month. One credit covers one candidate sitting, whatever proctoring is on, and unproctored tests on paid plans use no credits at all.
CodeSignal's pricing page lists no free plan for hiring; its free tier is for individual learners. TunnelQuiz has a free plan with 50 credits every month, with whole-screen capture and the tab, fullscreen and copy-paste checks every plan includes, so you can run a real round before paying.
Every row below is a capability one or both platforms publish. A dash means we found no public statement that the platform offers it.
| Feature | TunnelQuiz | CodeSignal |
|---|---|---|
| Questions and tests8 | ||
| Multiple-choice questions | Yes. Single and multiple answer, with negative marking | Yes. Quiz questions |
| Written answers | Yes. Marked by you, or AI-drafted on Pro | Yes. Writing questions |
| Coding questions | Partly, with conditions. 19 languages, AI-graded on Pro; code is not run | Yes. Runs code in a built-in IDE |
| Assessments that use AI tools | No. | Yes. Agentic assessments with Claude, Cursor and Codex |
| Test library | No. | Yes. 2,000 questions on Build; full certified library on Pro |
| Write your own questions | Yes. Every plan; CSV import up to 500 rows | Partly, with conditions. Pro plan, or with the Authoring with Cosmo add-on |
| AI interviews | Yes. Text chat interviews, Pro plan | Yes. AI Interviewers on every plan; more roles on Grow and Pro |
| Live technical interviews | No. | Yes. Every plan |
| Proctoring and test integrity9 | ||
| Captures of the candidate's screen | Yes. Whole screen every 60 seconds, every plan | Yes. Screen, video and audio recorded for the whole session |
| Webcam monitoring | Yes. Snapshots every 60 seconds, Essential and up; multiple-face flag on Pro | Yes. Continuous video recording |
| Identity check before the test | Partly, with conditions. Email one-time code on every plan; SMS code and check-in photo from Essential; no ID document check | Yes. Government photo ID matched to the person on camera, every plan |
| Copy and paste | Yes. Blocked and logged | Yes. Paste activity analyzed; can be disabled |
| Plagiarism and AI-use signals in answers | No. | Yes. Suspicion Score: solution similarity, telemetry, copy-paste |
| AI-assistant browser extensions | Yes. Known AI extensions flagged, Essential and up | Not stated. |
| Second display detection | Yes. In Chrome and Edge, every plan | Not stated. |
| Who reviews the flags | Automatic flags and a risk score; you decide, with optional auto-terminate from Essential | AI analysis, then CodeSignal's human reviewers validate flagged activity |
| Live view of candidates mid-test | Yes. Live monitoring from Essential, Live Follow on Pro | Partly, with conditions. Live interview proctoring on Pro |
| Team and integrations5 | ||
| ATS integrations | No. | Yes. Gem and Ashby on Grow; Workday, iCIMS, Oracle, Greenhouse and Lever on Pro |
| API | No. Webhooks on Business | Yes. Basic API on Grow, advanced API and webhooks on Pro |
| Single sign-on | Partly, with conditions. Google and Microsoft sign-in | Yes. SSO and SCIM on Pro |
| Your own branding | Yes. Pro plan | Yes. Every plan |
| Shareable, revocable report links | Yes. Essential and up | Not stated. |
| Pricing and access5 | ||
| Free plan for hiring | Yes. 50 credits every month | No. None listed; CodeSignal Learn is free for individuals |
| Entry price | $34/mo billed yearly, 150 credits a month | Build: $79/mo billed annually for 60 credits a year, or $99/mo for 5 credits a month |
| Extra candidates | Top-ups at $0.38 a credit or less | $20 per on-demand credit |
| Currency | USD, no tax added | USD |
| Pricing model | Flat plans: Free $0, Essential $34/mo and Pro $87/mo, both billed yearly (12% off monthly), Business quoted - 50/150/250 credits a month; unproctored tests free on paid plans, AI generation, grading and chat interviews metered separately in AI credits; in India, ₹1,319 and ₹3,519 plus GST | Credit plans: a credit is used when a candidate starts an assessment or interview; Grow from $479/mo billed annually, Pro quoted |
A proctored campus drive of 400 candidates uses 400 TunnelQuiz credits, about three months of the Essential plan's allowance, or a top-up. On CodeSignal's Build plan the same drive would use the year's 60 credits and 340 on-demand credits at $20 each, on its published pricing.
A proctored drive of 400 candidates
The Build plan's credits for the year, then on-demand credits
About three months of Essential's allowance, or a top-up
A lot of screening is aptitude, language, domain knowledge and judgment, asked as MCQs and short written answers. That is what TunnelQuiz is built around, with negative marking, per-question timers and shuffled question pools. CodeSignal covers business roles too, but its go-to-market and HR assessments start on Grow.
Tests for roles outside engineering
Where its go-to-market and HR assessments start
Negative marking, per-question timers and shuffled question pools
CodeSignal records the session and has it reviewed by AI and then its own reviewers. TunnelQuiz's live monitoring, from Essential, lists every candidate with their progress and flags while the test is still running, and you review the screen captures and flags yourself.
Who reviews the session
The session is recorded and reviewed for you
Live monitoring from Essential; the captures and flags are yours to review
It depends on what you are testing. For software engineering hiring, CodeSignal is the better tool: it runs code, offers agentic and live technical interviews, and ships a large certified question library. TunnelQuiz is built for proctored MCQ and written tests for any role. If most of your candidates are not developers, it is simpler and far cheaper per candidate.
As of October 2026, CodeSignal's hiring plans are Build at $79 a month billed annually (60 credits a year) or $99 a month billed monthly (5 credits a month), Grow at $479 a month billed annually (420 credits a year) or $599 monthly (35 a month), and Pro at a custom price. A credit is used when a candidate starts an assessment, AI interview or live interview, and extra credits cost $20 each. Prices are in US dollars; check CodeSignal's pricing page for current figures.
As of October 2026, TunnelQuiz has four plans: Free ($0, 50 credits a month that lapse at month end), Essential ($34 a month billed yearly, 150 credits a month, 5 seats), Pro ($87 a month billed yearly, 250 credits and 10 AI credits a month, 20 seats) and Business (quoted). Yearly billing is 12% off the monthly rate, and monthly billing is also available. Prices are in US dollars with no tax added. In India the same plans are ₹1,319 and ₹3,519 a month billed yearly, plus 18% GST. One credit is one candidate sitting, whatever proctoring is on. Unproctored tests are free on paid plans, paid credits roll over while you stay subscribed, and top-ups cost $0.38 a credit or less.
Yes, on every plan. Its AI proctoring records the candidate's video, audio and screen for the whole session, and candidates show a government-issued photo ID that is matched to the person on camera. Flagged activity is validated by human reviewers. CodeSignal's candidate help article says this proctoring data is reviewed by CodeSignal only, deleted within 15 days and not shared with the company that requested the report. TunnelQuiz works the other way around: you see the screen captures, snapshots and flags yourself, but there is no ID document check.
Yes, on the Pro plan, in 19 languages. An AI grades the answer against your question; the code is not compiled or run, and there is no IDE or project-style task. For deep technical assessment, CodeSignal is the purpose-built option. A sensible split is TunnelQuiz for aptitude and fundamentals, then a coding platform for the shortlist.
CodeSignal, from its Grow plan (Gem and Ashby) and on Pro (Workday, iCIMS, Oracle, Greenhouse, Lever and others). TunnelQuiz has no ATS integrations or public API; webhooks are available on its Business plan.
As of October 2026, and per candidate, by a long way. TunnelQuiz's Essential plan is $34 a month billed yearly for 150 credits a month, while CodeSignal's Build plan is US$79 a month billed annually for 60 credits a year. It is not a like-for-like comparison for engineering hiring, though, since CodeSignal runs code, offers agentic assessments and live technical interviews, and TunnelQuiz does not.
Start on the free plan with 50 credits every month. Screen captures on every plan, webcam snapshots and live monitoring from Essential.