Appearance
Fluency
A score per competency, per person, per team. Not vibes — built from a static catalogue of 11 chat-observable indicators applied to every Enterprise-pathway session in a rolling 30-day window.
This is the 3D fluency model — three competencies that show up in real chat traffic, plus a fourth that can't be observed there.
| Competency | What it measures |
|---|---|
| Description | How clearly the prompt frames the task — context, constraints, acceptance criteria |
| Delegation | How well the engineer hands off versus over-specifying or babysitting the model |
| Discernment | How critically the engineer evaluates the response before accepting it |
| Diligence | Not chat-observable — measured separately or omitted entirely |
The Diligence tile renders muted on the overview because there's no defensible way to score it from chat alone. We'd rather show "no signal here" than fake one.
What you see on the overview
The page is at AI → Insights → Fluency with sub-routes for /people and /teams.
Top strip — four competency tiles. Each shows the weighted score for the org over the last 30 days, the unweighted score, and a one-line summary. The Diligence tile shows only the summary.
Trend chart — a 12-week line chart with one series per observable competency, weighted scores in percent. Use this to spot whether your org is climbing or sliding.
Indicator prevalence table — every indicator in the catalogue, with:
- Org — your prevalence over the window.
- Baseline — the catalogue's reference prevalence (what "average" looks like in our reference dataset).
- Delta — your value minus the baseline. Green for above, red for below.
- Present / N — the absolute count of conversations where the indicator was present, over the count of conversations evaluated.
Indicators are colour-tagged by competency — Description (green), Delegation (indigo), Discernment (pink) — so you can read the table at a glance.
People and teams sub-routes
The /people sub-route ranks individuals by competency. The /teams sub-route does the same for teams. Both apply the same 30-day window and the same indicator catalogue, just sliced differently.
These pages are what an engineering manager opens before a 1:1 to spot whether someone's delegation has shifted, or what a director uses to compare two teams' adoption depth.
How a session is scored
Every Enterprise-pathway session that gets ingested is run through the classifier. The classifier produces:
- A boolean per indicator — present or absent in this conversation.
- A summary block with rationale (used in the per-session drilldown).
Org-level prevalence is then presentCount / conversationCount per indicator. Competency scores are aggregated across their indicators with a per-indicator weight from the catalogue.
Limits of the model
Worth being honest:
- Telemetry pathway orgs see no Fluency data — no prompt body, nothing to classify. Pick the Enterprise pathway if you want this.
- The classifier evaluates patterns in observable chat. It doesn't know whether the resulting code shipped, passed review, or caused an incident.
- Indicators are catalogue items, not custom prompts. We'd rather give you a stable benchmark than let every org drift its own.
Coding sessions only, today. The 3D model and its indicator catalogue are calibrated against coding conversations — the indicators look for things like "specifies acceptance criteria" or "asks a clarifying question after a vague response," which read clean against an engineering session and noisier against, say, a marketing-copy or image-prompt session. Modality-aware scoring for image, video, audio and chat is on the roadmap. Until then, Fluency is silent on non-coding traffic — we'd rather show no score than a misleading one. See Catalog for what coverage your org has across modalities today.
What you do from here
- Spot training opportunities. A team with low Description prevalence usually benefits from a prompt-quality workshop.
- Catch model regressions. A sudden drop in Discernment across the org often correlates with a model release that confidently hallucinates more than its predecessor.
- Calibrate hiring. Compare new joiners' first-30-day Fluency to the org baseline.
Empty state
If no Enterprise-pathway sessions have been captured yet, the indicator table renders an empty card asking you to enable the Enterprise pathway in Settings → AI → Agent Policy.