# Merge

> Merge is an AI-native code review assessment platform that has candidates review real pull requests while an AI agent simulates a real engineer responding to their feedback in real time.

Merge is an AI-native technical assessment platform built specifically for evaluating how engineers handle code review — a core skill that has grown in importance as AI-assisted coding becomes standard. Candidates work through a realistic pull request review loop, and Merge's AI agent responds to their comments in real time, simulating the back-and-forth of a real engineering team. The platform is currently in pre-launch, with access available by booking a demo.

## What It Is

Merge is a hiring assessment tool that replaces algorithm quizzes with a workflow that mirrors actual engineering work: reviewing a pull request, leaving comments on bugs, refactors, and security risks, and iterating on revisions as an AI agent addresses feedback. The result is a scorecard that hiring teams can use to evaluate how candidates reason, prioritize, and improve code — not just whether they can solve a contrived puzzle.

## How the Review Loop Works

The assessment follows a structured, multi-step loop:

1. **Read the codebase** — Candidates enter a small, realistic codebase with a scoped pull request.
2. **Review the PR** — Candidates comment on correctness bugs, refactors, and vulnerabilities.
3. **Submit the review** — Comments are evaluated for priorities before the AI generates a response.
4. **AI publishes a revision** — Merge addresses candidate comments and opens a fresh PR revision.
5. **Repeat** — The loop continues until time expires or the candidate approves the code.

This structure produces evidence of how an engineer thinks across multiple iterations, not just a single snapshot.

## Custom Assessment Configuration

Hiring teams can calibrate every assessment to the specific role they are filling. Configuration options include:

- **Difficulty** — Intern, New Graduate, Junior, Mid-level, Senior, Staff, or Principal
- **Specialization** — Frontend, backend, infrastructure, security, platform, distributed systems, data pipelines, and more
- **Languages** — Restrict to specific languages for depth, or allow broad stacks for generalist hiring loops

## Token Use and Efficiency Signals

Merge surfaces a signal that the vendor describes as unique to the platform: how efficiently a candidate uses AI tokens during the assessment. The reporting dashboard shows token use, estimated cost, and how candidates turn PR feedback into revisions. This is positioned as a direct response to the reality that engineers now routinely work with AI coding tools, and hiring teams need a way to assess that skill.

## Reporting and Hiring Signals

Every completed assessment produces a scorecard that connects candidate comments to code quality, risk detection, revision judgment, and practical hiring recommendations. The report is designed to give hiring teams something concrete to discuss rather than a raw algorithmic score.

## Current Status

Merge launched on ProductHunt and is currently offering pre-launch access via demo booking. The platform is built by engineers and is positioned as an early-stage product entering the technical hiring market.

## Features
- AI-simulated PR review loop
- Real pull request assessments
- Custom difficulty levels (Intern to Principal)
- Specialization configuration (frontend, backend, infra, security, etc.)
- Language restriction settings
- Token use and efficiency tracking
- Estimated AI cost reporting
- Candidate scorecard with hiring recommendations
- Code quality and risk detection signals
- Revision judgment evaluation
- Real-time AI agent responses to candidate comments

## Platforms
WEB, API

## Pricing
Paid

## Links
- Website: https://mergeoa.com
- Documentation: https://docs.mergeoa.com/
- EveryDev.ai: https://www.everydev.ai/tools/merge-oa