AI agent for training coordinators
Instructor Quality Observation Agent
Instructors receive specific feedback and a coaching focus, with change checked in later sessions.
What it does
An instructor scores 4.6 out of 5 but learners cannot apply the skills, and the average hides what to improve. This agent reads session feedback, observation notes and learner results for each instructor and looks for patterns, such as high satisfaction but low quiz scores in the same topic. It proposes a coaching focus and, after the next sessions, checks whether the pattern changed. If it did not, it suggests a different approach. It compares what learners say with what learners can actually do, so a popular but weak session is noticed. Edge case: one bad session with a broken projector is explained and excluded. The coordinator approves feedback.
How it works
Follow the arrows from top to bottom. The orange dashed arrow is the loop: when a check fails, the agent goes back and tries again.
Read the steps as a list
- Review period ends
- Read feedback, observation notes and learner results by instructor
- Find patterns across sessions
- Is each pattern repeated and not caused by a one-off event?If not: exclude the one-off and recompute. Back to step 3.
- Propose a coaching focus for each instructor
- Coordinator approves the feedbackThe agent waits here for your OK.
- Share approved feedback with the instructor
- Read the next sessions' results
- Did the pattern improve?If not: propose a different coaching approach. Back to step 5.
- Coaching report
How it decides
It looks for repeated patterns across sessions and ignores single sessions with a known cause.
- A pattern needs at least 3 sessions.
- Exclude sessions with technical or room problems.
- Compare feedback with learner results, not scores alone.
- Coaching suggestions are specific and limited to two per instructor.
Make it yours
Every agent is a starting point. You choose these settings for your own situation.
- Pattern minimum (default 3 sessions)
- Data sources
- Review period
- Coaching limit
- Feedback format
What keeps you in control
It always asks you first
- All feedback before it is given to an instructor
- Any performance conclusion
Hard limits
- Do not rank instructors against each other in feedback.
- Do not share feedback without approval.
It stops when
- Done: feedback is delivered and follow-up is checked
- Stop: too little data, so the agent waits for more sessions
Set it up
We guide you through the set-up, step by step
Members get the full set-up guide for this agent. No technical skills needed: you copy, paste and upload.
- One set of instructions to paste into your AI, with the clicks for ChatGPT, Claude, Microsoft 365 Copilot, Gemini and Grok
- The agent then walks you through connecting your own data, one source at a time
- A downloadable copy with the flow chart, the rules and the full guide
