Developer Insights
Who is using AI coding tools, on what, and where the cost and time go, across laptops, cloud workspaces and automated agents. The evidence engineering leaders need.
Beta · Commercial licence · Works with Claude Code
Cost and effort
Every session is costed and timed, so you can see what typical work costs, what a commit costs, and which sessions are worth a closer look.
Developers and agents
Each person’s sessions, cost and output sit beside the team’s typical figures, never in a ranking. Runs started by automation get the same profile, so you can see what an AI reviewer or fixer costs and produces.
Sessions
Open a session to see each prompt, the steps AI took, the files it touched and what each turn cost, in a terminal or chat view. Laptop sessions arrive through the Orchestrated plugin; cloud sessions archive themselves.
Skills, models and repos
Find the team skills that get used and the ones that don’t, what each costs and the commits they produce, and where the money goes by type of work, AI model and repository.
Projects
Group the repositories behind one piece of work, such as a front end, an API and its infrastructure, and see its cost, output and people in one place.
Who is using AI coding tools, day by day.
Laptop, cloud workspace and automated agent sessions, counted separately.
Typical and outlier costs, with the sessions behind them.
What AI-assisted output actually costs.
Which team skills get used, by whom, and what they produce.
Where the money goes by AI model and type of work.
Cost, activity, skills and people for every repository and project.
Each person and each automated agent, compared with the team typical.
The prompts, tool calls and file changes behind any session you can see.
The Orchestrated plugin backs up laptop sessions in the background, with nothing to configure.
Developers choose which laptop sessions their organisation can read.
How often web access, network calls and commands fail inside cloud workspaces.
DORA metrics alongside AI usage, so you can see how AI changes delivery speed and stability.
How agentic each team's way of working is, tracked over time, so coaching is backed by data.
Which prompts get the work done first time, and which need repeated attempts.
Sessions where AI goes round in circles: retries, repeated attempts and dead ends.
Developer Insights
Connect your teams’ AI coding tools and see adoption, cost, skills and output in one place. Licensed and supported by our team.
Developer Insights + coaching
Our coaches use what Developer Insights shows to target where teams need help, then track the uplift as AI-native practice takes hold.
More on engineering coachingAI performance
In productionReal-time performance observability for the AI you run in production.
AI enablement
In productionThe right AI skills on every machine, and a safe way to ship what people build.
Compliance and risk
In pilotA System One AI model, built specifically for compliance, risk and audit teams.
Product delivery
In developmentWhere product owners, stakeholders and AI agents shape and ship software together.
Documentation
In developmentDocumentation that writes itself, and keeps itself up to date.
Request access to the beta and we’ll help you connect your first teams.