Answers
Questions worth asking about AI engineering.
AI agents have changed how engineers spend their day. These are the questions people ask about that change, answered directly: working with agents, attention, rework, privacy, and whether it is helping.
For engineers
Working with AI agents
Supervision, concurrency, helper agents, models and routes.
- How can I see how much time I spend supervising AI agents?Flow splits your AI-assisted time into hands-on work and supervising agents, from the history your tools already keep, so you can see where your attention goes.
- Is running more AI agents at once actually helping me?Flow shows how many agent sessions you had in flight, how often they needed you, and what failed, so you can compare your heavier-concurrency days with your lighter ones.
- How do I measure my Claude Code, Codex or Cursor use beyond tokens?Tokens show what the model consumed. Flow reads Claude Code, Codex, Cursor and Gemini CLI history to show how you worked with agents: steering, switching and rework.
- Can my coding agent read my own work history?Flow exposes two local MCP tools so a coding agent can read your own sessions and days: timings, tools, models, supervising time, failures and compactions. Local only.
- How can I see my AI coding work split by model vendor and route?Flow shows your own AI coding work split by model vendor and by how it was served: direct, through a cloud provider, a gateway, a custom endpoint or on device.
- Does a different model or effort level change how my AI coding goes?Flow can assign which model or effort level you use on each working day, in pairs, then lay the failed tool calls side by side, so you compare on your own work.
- Where does my Claude Code quota go?Flow reads the model, effort level and permission mode of each Claude Code response, and keeps your weekly and session limit notices, so you can see what uses your quota.
Attention and workload
Interruptions, switching, after-hours work and how the week lands.
Failures and rework
Failed tool calls, re-prompts and work you go back and redo.
- How can I tell whether AI is creating more rework for me?Flow counts failed tool calls, re-prompts and retry storms from your AI tools' history, so you can see whether rework is rising and when it happens.
- How often do I redo the work my helper agents hand back?For each Claude Code helper agent, Flow records how long it ran, its tool calls and failures, what it handed back, and whether you went back and edited its files.
Whether AI is helping
Baselines, declared changes and readouts beyond tokens and commits.
Privacy and trust
What is read, what is sent, and who can see it.
For engineering leaders
Working with AI agents
Supervision, concurrency, helper agents, models and routes.
Attention and workload
Interruptions, switching, after-hours work and how the week lands.
Whether AI is helping
Baselines, declared changes and readouts beyond tokens and commits.
- What tools measure how engineers work with AI coding agents?Ascenda measures how engineers work with AI coding agents: steering, decisions, interruptions and rework, reported to organisations only as aggregates of ten or more.
- How do I know whether rolling out Claude Code (or another AI agent) helped my team?Declare the change, the measure and the window before you roll out. Ascenda freezes the baseline and reports the readout whichever way it goes.
- Can we measure how our engineers work with AI from history we already have?Ascenda's 30-day Report works from the history engineers' AI tools already keep, for a cohort of ten or more, with no integration, and ends in a written deliverable.
- What should engineering managers measure when rolling out AI coding agents?Measure adoption, the human work agents create, friction and rework, how engineers say the work feels, and delivery outcomes, against a baseline set before the rollout.
- How do we test whether switching AI model or vendor helped our engineers?Roll the switch out to groups in turn. Ascenda compares failed tool calls per active hour with groups that haven't switched yet, and says when it's too small to tell.
Privacy and trust
What is read, what is sent, and who can see it.
See it on your own work
Flow is a free macOS app that reads the AI tools you already use. For a team, the 30-day Report works from imported history for a cohort of ten or more, with no integration, and the 90-day Study runs live.