AI-powered reviewBeta · Free to try

Review take-home assignments in 60 seconds.

Stop spending 15 minutes per repo. Drop the assignment brief and GitHub URLs — CodeVerdict maps every requirement to code, detects AI shortcuts, runs the app, and hands you a scored report with interview questions ready.

60s*
avg. review time
100%*
requirements tracked
0*
lines missed

* Based on typical assessments during beta testing.

PM
Priya Menon
priya-m/ship-dash
91
/ 100
Strong hire11/12 requirements8% AI
Requirements met92%
Code quality88%
Test coverage86%
AI-written code8%
AI-generated interview questions
"Your websocket reconnect logic uses exponential backoff capped at 30s — how would you handle server-forced disconnects differently?"
Report ready in 58s
The problem

Take-home reviews are broken for everyone.

Senior devs lose 2–3 hrs/day
Reviewing 10 candidates means 2+ hours of cloning, running, and reading code that may not even build.
Companies see only the README
Most reviewers skim the docs, miss key requirements buried in code, and reject great engineers.
Candidates get zero feedback
After hours of work, a silent rejection. No insight into why — even when they mostly got it right.
The solution

One brief. Any number of repos. One minute.

Requirement-level proof
Every requirement is mapped to the exact file and line that satisfies — or fails — it.
AI-code detection, with receipts
Per-file analysis flags likely AI-written sections so you know if the candidate actually wrote this.
Live sandbox execution
We boot every repo in an isolated VM, call the APIs, and confirm it actually runs.
Reject or invite — with one click
Auto-draft a rejection email or send interview questions directly from the report.
Everything in one place

Built for hiring teams who take code seriously.

Requirement tracking
Upload any PDF, Markdown, or DOCX brief. We extract every requirement and check each one against the codebase.
AI-code detection
Token-level perplexity, naming entropy, and commit history analysis tell you if the work is genuine.
Sandbox execution
Repos boot on ephemeral e2b VMs. We run build scripts, start the server, and call your endpoints live.
Chat with code
Ask questions grounded in the actual repo. "Where is rate limiting?" returns the exact file and line.
Candidate comparison
Compare up to 4 candidates side by side across every dimension — like a product comparison table.
One-click emails
Send polished rejection emails or interview invites with tailored questions — straight from the report.
Dev timeline
Commit history visualised. See when work was done, velocity, and whether code was written under pressure.
Custom scoring
Set your own weightings — requirements, quality, tests, security. The score reflects what matters to you.
No signup to start
Paste a brief and GitHub URLs. Get reports. Create an account later to sync and invite team members.
Frequently asked questions

Common questions about CodeVerdict

What is CodeVerdict?

CodeVerdict is an AI-powered platform that reviews developer take-home coding assignments in under a minute. You drop the assignment brief and a GitHub URL (or zip); CodeVerdict maps every requirement to the code, detects AI-written shortcuts, runs the project in a sandbox, and produces a scored report with tailored interview questions.

How long does it take to grade a take-home assignment?

Most assignments are reviewed in 60 seconds or less. The agent reads the brief, clones the repo, runs setup / tests / execution in an E2B sandbox, and writes the report in parallel. Larger repos with heavy dependencies may take up to two minutes.

What can I submit — GitHub repos, zip files, or both?

Both. You can paste one or more GitHub repository URLs (public or private with a GitHub token) or upload a zip file of the candidate's solution. CodeVerdict handles each the same way: clone, install, run, evaluate.

Do I need an account to try it?

Yes — a free CodeVerdict account is required. Sign up takes ~10 seconds (Google/GitHub one-click) and we never share your data. Accounts let your assessments sync across devices and your team see the same dashboard.

How does CodeVerdict detect AI-written code?

It combines three signals: token-level perplexity (LLM-written code has unusually uniform probability distributions), naming and structural entropy (AI tends to use overly consistent patterns), and commit-history analysis (sudden large commits with no incremental work). The output is a calibrated 0–100 AI score, not a black-box verdict.

Is candidate code safe to run? Where does it execute?

Yes. Every repository runs inside an isolated E2B sandbox — a single-use virtual machine that is destroyed immediately after the report is generated. The candidate's code never touches your machine or CodeVerdict's production servers.

Which programming languages and stacks are supported?

CodeVerdict supports Node / TypeScript, Python, Go, static front-ends, and any Docker-based project out of the box. The agent detects the stack automatically from project hints (package.json, requirements.txt, go.mod, Dockerfile, etc.) and picks the right setup path.

Can I customise how candidates are scored?

Yes. The displayed score is computed client-side from your weightings — requirements met, code quality, test coverage, security, AI-written code. Adjust the weights in Settings and the verdict (Strong hire / Hire / No hire) updates instantly across every submission.

Ready to cut your review time by 90%?

Start reviewing take-home assignments in 30 seconds.