Skip to content

Pricing

Free for small teams, one price after that

Every plan includes all repos, the adoption ladder, the guardrail checklist, teams and nightly scans. The paid plan, for orgs with more than 10 engineers, adds Impact analysis, which is in design.

Free

$0

  • Up to 10 engineers
  • The ladder, guardrails, teams and nightly scans
  • No card needed

Paid

$10 per engineer per month

  • For orgs with more than 10 engineers
  • Everything in Free
  • Impact analysis In design
  • The first 10 engineers stay free
  • 14-day trial before the first charge
  • Billed monthly in US dollars; cancel at any time

Impact analysis: is AI adoption paying off?

In design

Impact analysis uses your organization's merged pull-request history, the same read-only GitHub data Chamlis already scans, to answer one question: did AI adoption make delivery faster without making it riskier? It compares pull requests co-authored with an AI coding agent against the rest, after adjusting for the size of the change, the type of change and the repository, and reports a speed figure next to risk figures, each with a plausible range.

  • For: A VP of Engineering or CTO at an organization with more than 10 engineers who has to show that AI adoption is paying off without eroding quality.
  • Not for: Solo maintainers and small teams. They usually have too few merged pull requests for a reliable answer.

What one analysis found

One engineering organization, about 2,000 merged pull requests over 90 days. A little over half were co-authored with an AI coding agent.

  • SpeedNo detectable change in time to merge. The estimate is 6% faster, with a plausible range from 27% faster to 15% slower.
  • Why the adjustment mattersA raw comparison said co-authored pull requests took more than twice as long to merge. After adjusting for size, type of change and repository, the gap disappeared: co-authored pull requests were larger and sat in different repositories.
  • QualityCo-authored pull requests failed checks after merge about 4 percentage points less often (range 1 to 7 points), with no detectable difference in follow-up fixes within 48 hours.
  • The guardrail to watchCo-authored pull requests were merged without a human approval about 10 percentage points more often (range 6 to 14 points).
  • Where the time goesOn reviewed pull requests, about 43% of the time from open to merge was spent waiting for a first review, and about 40% in draft.

Read these figures with care

  • Figures are rounded.
  • They are associations after adjustment, not proof of cause.
  • They come from one organization and will differ for yours.
  • No names, code or comment text were used.

What it uses

  • Organization and team level only, never per person.
  • Built from pull-request metadata, not code, prompts or comment text.
  • An example or a result is never shown outside your organization without your say-so.

Today the analysis is run by hand on Lighty's analysis engine, and the view in the dashboard is in design. Talk to us if you want it for your organization.

What an org pays

EngineersPer month
10$0
11$10
25$150
50$400
100$900

Who counts as an engineer

An engineer is an org member with at least one commit or pull request in the last 90 days. Bots and outside contributors are not counted. The count is taken again by each full scan, and each invoice uses the latest count, with no charge or credit for changes partway through a month.

If a free org grows past 10 engineers

The dashboard pauses until an org owner starts the paid plan. Scans keep running and nothing is lost, so the results are there as soon as the plan starts.