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 follow-up 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 about 10 seconds via Google one-click. 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.

HackerRank vs Real World Projects for Technical Interviews

The Great Divide in Technical Hiring

When it comes to screening software engineers, the industry is largely split into two camps:

The Algorithmic Camp: Platforms like HackerRank, LeetCode, and CodeSignal that test data structures, algorithms, and logical puzzles.
The Real-World Camp: Take-home projects, paid trials, and repository reviews that test a developer's ability to build actual applications.

Which one is better for your company?

The Case for HackerRank (Algorithmic Puzzles)

HackerRank and similar platforms have been the industry standard for a decade, heavily popularized by FAANG companies.

Pros:

Highly Scalable: You can screen 10,000 candidates with zero human engineering time.
Objective Baselines: It provides a clear, numerical score based on test cases passed and execution time.
Tests Fundamentals: It ensures the candidate understands Big-O notation, memory management, and core computer science concepts.

Cons:

False Negatives: Many excellent, senior product engineers will fail a HackerRank test because they haven't inverted a binary tree since college.
High Anxiety: It creates an unnatural, high-pressure environment.
Poor Predictor of Day-to-Day Work: Knowing how to write a highly optimized sorting algorithm rarely translates to building a scalable React frontend or a maintainable CRUD API.

The Case for Real-World Projects

Real-world projects ask candidates to do what they will actually be doing on the job: pulling an API, rendering UI, writing unit tests, and structuring a small application.

Pros:

High Predictive Validity: If they can build a great React app in the interview, they can probably build a great React app for your company.
Tests the "Whole" Engineer: It evaluates code organization, testing habits, documentation (READMEs), and architectural choices.
Better Candidate Experience: Senior engineers vastly prefer building things over solving abstract puzzles.

Cons:

Hard to Grade at Scale: It takes a human engineer 30-60 minutes to properly review a take-home project.
Time Commitment: It asks a lot of the candidate's free time.

The Verdict: Context is Everything

If you are Google hiring 5,000 new grads a year to write low-level C++ infrastructure, algorithmic puzzles are necessary to manage the funnel.

If you are a startup or mid-market company hiring 10 developers to build a SaaS product, Real-World Projects are vastly superior. You need product-builders, not puzzle-solvers.

Bridging the Gap: Automated Real-World Grading

The only reason companies default to HackerRank is that grading real-world projects is too slow.

But what if you could have the scalability of HackerRank with the predictive validity of a real-world project?

Enter Codeverdict.

Codeverdict allows you to issue real-world, repository-based take-home assignments and automates the grading process. It evaluates architecture, code quality, and functional correctness instantly, giving you the best of both worlds.

Ready to see how Codeverdict can automate your take-home grading? Try Codeverdict today.