Energy-aware tutoring is a system where the tutor knows the student's current energy state and adapts the session — length, difficulty, intensity — accordingly. aitutors.me uses a GREEN/AMBER/RED traffic light, set by the student or parent at the start of each week. This guide explains how it works and why it makes a measurable difference.
The problem it solves
A traditional tutoring session is booked at a fixed time. The tutor arrives ready to push. The student arrives at one of four wildly different states:
- Fresh on a Sunday afternoon: ready for hard work
- Tired Tuesday after a maths test that day: mid-capacity
- Wrecked Wednesday after football and homework: low capacity
- Fried Friday in mock-exam week: below working threshold
The tutor without state awareness runs the same session in all four. Three of those four sessions waste most of the cost.
The four states
| State | Indicator | Tutor adapts by |
|---|---|---|
| 🟢 GREEN | Slept well, low-stress week, free time available | Full session, challenge problems, push boundaries |
| NORMAL | Average week, mild fatigue | Standard session, balanced difficulty |
| 🟠 AMBER | Two big commitments, tired, mid-week dip | Shorter session (15–20 min), core practice only |
| 🔴 RED | Exam week pressure, ill, overloaded | Suggest rest. No new content. OK to skip entirely. |
How it works in practice
Step 1: Parent or student sets the weekly state
At the start of each week:
Mentor: "How's the week looking?"
Student: "Football Mon and Wed, mock Friday — busy."
Mentor: "OK — I'll set AMBER for the week. We'll do shorter
sessions, lots of review, nothing new. Sound right?"
Student: "Yeah."
Step 2: Mood check before each session
Mentor: "Quick energy check — 1–3, how are you tonight?"
Student: "2."
Mentor: "OK, 15-minute session. Routing to Pi."
Step 3: The subject tutor receives the state
Pi gets told: "Today is AMBER-2. Keep it light." Pi then chooses:
- Easier problems
- Shorter sessions (timer set internally)
- More celebration of effort
- No "challenge mode"
Step 4: At RED, the tutor refuses
If the state is RED, Pi will say something like:
"Looks like a hard week. Let's skip tonight. A 10-minute walk or some sleep will do more for tomorrow's maths than this would."
This is non-negotiable — Pi will not run a full session in RED. The Mentor agent owns the gate.
Why it matters (the wellbeing case)
UK guidance (Royal College of Paediatrics) recommends 9–11 hours of sleep for ages 11–14. Below 8 hours, working memory drops measurably. A tutoring session run on a tired brain wastes the cost; one run after rest doubles the value.
The system pays for itself in sleep alone.
Why it matters (the learning case)
Adaptation isn't just kindness — it's pedagogy. Learning consolidates during sleep. Pushing tired students into hard new material associates the topic with frustration, which slows retrieval next time. Better to do review in AMBER and save new topics for GREEN.
The two failures it prevents
Failure 1: The Wednesday-night burnout
Without an energy system, a child does Wednesday's tutoring after football, struggles, gets frustrated, and starts disliking maths. Over weeks, this becomes "I'm bad at maths" — which is wrong, they were just tired.
Failure 2: The exam-week panic
In exam week, well-meaning parents push more tutoring. With an energy system, the tutor sees RED and refuses, redirecting to rest and review. Parents push back; the tutor holds the line.
Where this came from
aitutors.me's energy system grew out of practical experience with one student (the founder's daughter Rina — see the founder story). She was over-scheduled in a way that defeated traditional tutoring. The energy system was the fix that made tutoring stick.
The methodology is documented at docs/methodology/Weekly-Energy-System.md in the open product repo.
How it's different from "adaptive learning"
Adaptive learning platforms (CENTURY, Sparx) adapt to a student's knowledge level — what topics they've mastered. Energy-aware tutoring adapts to their state — how they're doing today. Both are valuable; they solve different problems.
The future of AI tutoring will probably combine both: adaptation to level AND state. aitutors.me leans into state because no other player has.
How parents set the state
Parents can override the weekly state via the dashboard. Useful when:
- The child says GREEN but the parent knows it's been a rough week
- A planned commitment was added late (and the energy budget needs adjusting)
- Recovery is needed after illness
The student's self-report is the default; parent override is the backstop.
FAQ
What does energy-aware tutoring mean?
A system where the tutor knows the student's current weekly energy and adapts session length, difficulty, and intensity. aitutors.me uses GREEN/AMBER/RED.
What is the Mentor agent in aitutors.me?
Mentor is the gateway agent. It checks weekly energy and daily mood, routes to subject tutors, and refuses sessions on RED.
How is energy state determined?
Self-reported by student or parent at the start of each week, with re-check before each session. Self-report correlates well enough with actual learning capacity for tutoring purposes.
Related reading
Methodology by Jason at aitutors.me. Updated 20 May 2026.