AIQ Rank vs LinearB — AI coding agent proficiency vs engineering intelligence | AIQ Rank

TL;DR

LinearB is an engineering intelligence platform. It pulls data from GitHub, Jira, and CI to produce dashboards on cycle time, PR throughput, deploy frequency, and team health.

AIQ Rank is an AI coding agent proficiency benchmark. It reads Claude Code, Codex CLI, OpenCode, Cursor, and Cowork session logs to measure how well a developer or team actually uses AI tools.

If you want to know how fast your team ships, use LinearB. If you want to know how AI-fluent your team is, use AIQ Rank. They look at different layers of the engineering workflow.

What LinearB does

LinearB is a developer experience and engineering metrics platform. It connects to your VCS (GitHub, GitLab, Bitbucket), issue tracker (Jira, Linear), and CI/CD to produce metrics like cycle time, review depth, PR pickup latency, and deploy frequency. Teams use it to identify bottlenecks, run goal-setting on engineering velocity, and benchmark against industry data.

Strengths:

• Mature engineering-intelligence category, well-established with enterprise customers
• Broad metric coverage across the SDLC
• Strong reporting for VP and CTO audiences
• Newer features around AI tool ROI ("measuring Claude Code ROI" type analyses)

What it does not measure directly: the skill of the developer using AI coding agents. LinearB can show that PR throughput went up after Claude Code adoption, but it cannot tell you which developers are using Claude Code well versus which ones have it installed and never use plan mode, custom skills, or parallel agents.

What AIQ Rank does

AIQ Rank measures AI coding agent proficiency itself, not the downstream output. The plugin scans your Claude Code, Codex CLI, OpenCode, Cursor, and Cowork local session logs and produces a score across six dimensions:

• Customization (CLAUDE.md, hooks, custom commands)
• Parallel Agents (subagent orchestration)
• Tool Breadth (skills, MCP servers, plugins)
• Planning (plan mode, structured workflows)
• Multi-Tasking (parallel sessions)
• Custom Skills (reusable workflows the developer built)

For an engineering leader, this surfaces the question LinearB can't directly answer: who on my team is actually using AI well? It produces a private leaderboard of AI proficiency by developer, by team, by tool. Free during launch.

When to pick which

Pick LinearB if:

• You need engineering velocity and team-health dashboards
• You're managing PR cycle time, deploy frequency, and review depth
• You're an executive who wants engineering-output metrics for board reporting
• You want correlation analyses (e.g., AI adoption vs PR throughput)

Pick AIQ Rank if:

• You want to measure who on your team is actually fluent with Claude Code, Codex CLI, or OpenCode
• You're screening developer candidates for genuine AI tool proficiency
• You want adoption metrics for AI coding agents beyond installation counts and token usage
• You want a leaderboard for healthy team competition on AI skill growth

Many teams use both. LinearB at the output layer (did velocity improve?) and AIQ Rank at the input layer (do my engineers actually know how to use AI tools?).

Common questions

Doesn't LinearB already measure AI tool impact?

LinearB measures downstream metrics: cycle time, PR throughput, deploy frequency. It can correlate those with AI adoption, but it cannot tell you which specific developers are using AI well. AIQ Rank fills that gap by reading actual session activity.

Can AIQ Rank integrate with LinearB?

Not today. The two systems answer different questions. AIQ Rank focuses on the skill layer (reading transcripts and session logs); LinearB focuses on the SDLC layer (reading VCS and CI/CD data).

What about token-cost dashboards?

Both LinearB and the AI providers themselves (Anthropic Console, OpenAI Dashboard) report token spend. Neither is a proficiency measure. AIQ Rank's whole premise is that token counts are a misleading metric. Real activity, not counting tokens.

Is AIQ Rank an alternative to engineering-intelligence platforms?

No. It's a complement. AIQ Rank measures the AI tool skill of your developers. LinearB measures the output of your engineering org. Both are valid, and they answer different leadership questions.