Medhank

Built for institutions and hiring teams in India

Run a real test online, and trust the result.

Medhank is an online testing platform for institutions and companies. Build a question bank, schedule a sitting, proctor it as it happens, and publish scored results — in one place, with an audit trail behind every mark.

Accounts are created by your institution. There is no public sign-up.

How it works

From a blank question bank to a published result

Four steps, each one owned by someone different at the institution — and each one recorded.

  1. 01

    Build the bank

    Write questions once and version them. Multiple choice, written answers and code, each with its own marking rules.

  2. 02

    Shape the test

    Group questions into phases with their own timing and rules, then set who may sit it — a cohort, a list, or one invited candidate.

  3. 03

    Proctor the sitting

    Watch attempts as they run. Tab switches, disconnections and device changes are recorded against the attempt, not guessed at afterwards.

  4. 04

    Score and publish

    Automatic marking where the answer is objective, evaluators where it is not, and a breakdown a candidate can be shown when they query it.

Features

The parts that decide whether a result holds up

Everything here exists because the alternative is a spreadsheet and an argument.

Versioned question bank

A question edited after a test is live does not change what was asked. Every attempt is scored against the version the candidate actually saw.

Phases, not one long paper

A test is a sequence of phases, each with its own duration, navigation rules and question set — an aptitude round and a coding round in one sitting.

Proctoring that records, not accuses

Focus loss, disconnections, screen shares and device changes become a timeline on the attempt. A human decides what it means.

Survives a bad connection

Answers are saved locally and synced as the network allows, with a low-bandwidth path for candidates on weak mobile connections.

Scoring you can explain

Objective questions score themselves; written and code answers go to evaluators. Either way the breakdown is there when a candidate disputes it.

An audit trail by default

Who changed a question, who opened a sitting, who overrode a mark. Written as it happens, append-only, and readable when someone asks.

Who it's for

Two kinds of organisation, one platform

Both are running the same problem: ask a lot of people the same questions, fairly, and be able to defend the answer.

Institutions

Colleges, universities and training institutes.

  • Departments, cohorts and roll numbers, not a flat list of emails
  • Internal assessments and end-of-term papers on the same bank
  • Results a student can be shown, with the working behind them

Companies

Hiring teams screening at volume.

  • Send a test to one candidate or a thousand from the same screen
  • Coding questions run against real test cases, not eyeballed
  • A shortlist ordered by score, with every attempt still reviewable

Already have an account?

Sign in once here. We will take you to your console or to your tests, depending on what your institution has given you.