What’s Collected

Tandemu uses the OpenTelemetry standard to collect three types of data from Claude Code sessions:

Privacy

Tandemu is designed for stealth observability — it tracks session-level metrics, not individual actions:

What IS tracked:

• How long a session lasted
• How many lines of code were AI-generated vs manually typed
• Which files had repeated errors (friction)
• Deployment frequency and failure rates

What is NOT tracked:

• Individual keystrokes
• Screen recordings
• Prompt content (what the developer asked Claude)
• Code content (what was written)
• Idle time or break tracking

Pipeline

Developer's Terminal
       │
  Claude Code (emits OpenTelemetry data)
       │
       ▼
  OTel Collector (port 4317/4318)
  ├── Validates and batches data
  └── Tags with organization ID
       │
       ▼
  ClickHouse (analytical database)
  ├── otel_traces — session and deployment data
  ├── otel_metrics_sum — code line counts
  └── otel_logs — friction events
       │
       ▼
  NestJS Backend (queries ClickHouse)
       │
       ▼
  Dashboard / Claude Code Skills

Metrics Explained

AI vs Manual Ratio

Measures the proportion of code generated by Claude Code versus code typed by the developer.

• Tracked via the code.lines.ai_generated and code.lines.manual metrics
• Calculated per session and aggregated by team, sprint, or time period
• A ratio of 2.5x means 2.5 lines of AI code for every 1 line of manual code

Friction Heatmap

Identifies code areas where developers struggle repeatedly.

Friction is detected by:

• Prompt loops — The developer asks Claude to fix the same issue multiple times
• Tool execution errors — Claude’s file edits or commands fail repeatedly

Each friction event is tagged with the file path, so Tandemu can show which files cause the most trouble.

Severity levels:

• Critical (red) — 5+ prompt loops or 3+ errors
• Warning (yellow) — 2-4 prompt loops
• Info (green) — 1 prompt loop

DORA Metrics

The four DORA metrics are inferred from telemetry:

• Deployment Frequency — Counted from deployment trace events
• Lead Time for Changes — Duration between first commit and deployment
• Change Failure Rate — Percentage of deployments that result in failures
• Time to Restore — Duration of incident resolution traces

Passive Time Tracking

Session duration is calculated from trace start/end timestamps:

• A “session” starts when a developer begins a Claude Code session
• It ends when they close it or after a period of inactivity
• Total hours are aggregated per developer per day
• AI-assisted hours are sessions where AI tools were actively used

Data Retention

By default, telemetry data in ClickHouse has a 90-day TTL. Older data is automatically purged. This can be configured in the ClickHouse table definitions.