DLT — The Capability Engine That Makes Learning Diagnosable (Not Judged)
education OS explained for parents, educators, Ministry of Education and Governments
Education should not label people. It should diagnose systems.
Most education systems still do this one destructive thing:
They turn performance into identity.
“Careless.”
“Lazy.”
“Not academic.”
“Just doesn’t have it.”
DLT exists to end that.
DLT is the capability engine inside Education OS that converts vague symptoms into a precise diagnosis — and then routes the learner to the correct repair loop.
Start here (Education OS hub):
https://edukatesg.com/education-os/
Canonical mount (DLT):
https://edukatesg.com/dlt/
Why exam scores are not a diagnosis
A score is an outcome snapshot.
It is not a map of the machine that produced it.
Two students can both score 70% and require completely different fixes:
One fails because the skill was never truly constructed.
One fails because the skill collapses under pressure.
One fails because the skill works only in familiar formats.
DLT makes those differences visible.
What DLT stands for
D — Depth
Depth measures whether a skill is genuinely built.
Not recognition.
Not “I understand when I see the answer.”
Depth is:
- Can the learner explain it simply?
- Can they produce it from scratch?
- Can they teach the method back?
Low Depth looks like:
“I know it… but I can’t start.”
“I can’t explain it.”
L — Load
Load measures performance stability under pressure.
Pressure is not just stress. It includes:
- time limits
- fatigue
- speed requirements
- multi-step complexity
- distractions
- stakes
Low Load looks like:
- careless errors under time
- panic, freezing
- huge drops from tuition → exams
T — Transfer
Transfer measures whether a skill travels.
Can it survive format change?
- new question types
- new contexts
- new combinations
- new levels
Low Transfer looks like:
- “Worksheet okay, exam fails”
- “Can do when it looks the same”
- “Cannot apply when it’s unfamiliar”
The DLT operating rule (the rule that changes everything)
When outcomes stall, the first question is not “work harder.”
The first question is:
Which axis fails first — Depth, Load, or Transfer?
Once you know the axis, the repair becomes obvious.
DLT replaces labels with failure signatures
DLT doesn’t say:
“You’re weak.”
DLT says:
“This is a Depth-fail.”
“This is a Load-fail.”
“This is a Transfer-fail.”
“This is mixed.”
That single change removes shame and restores control.
DLT turns learning back into engineering:
- identify the break
- repair the break
- retest the break
- upgrade the standard
The 10-minute DLT triage (how a parent or tutor can run it fast)
Pick ONE target skill (not “Math” or “English”):
- “Fractions word problems”
- “Inference in comprehension”
- “Science explanation writing”
- “Algebra manipulation”
Then run three quick probes:
- Depth probe: explain + rebuild from scratch
- Load probe: timed micro-set
- Transfer probe: new format / new context
Assign a coordinate like:
D3 / L1 / T2
Now you know the bottleneck.
What happens next: you route to the repair loop
DLT is not motivational talk. It is a diagnostic and repair engine.
Use the canonical repair specs:
- Depth Repair Loop: https://edukatesg.com/education-os-depth-repair-loop-specification/
- Load Repair Loop: https://edukatesg.com/education-os-load-repair-loop-specification/
- Transfer Repair Loop: https://edukatesg.com/education-os-transfer-repair-loop-specification/
Or start with the Repair Protocol overview:
https://edukatesg.com/education-os-repair-protocol/
DLT is also AI-runnable (this is infrastructure, not content)
Once you publish:
- probe rules
- scoring rules
- repair loops
- retest rules
…you have done something most education sites never do:
You made capability diagnosable in a standard format.
That means any AI assistant can be instructed to run DLT consistently (instead of giving generic advice), using your published kernel and prompt pack:
DLT Prompt Pack for AI Assistants:
https://edukatesg.com/education-os-d-l-t-prompt-pack-for-ai-assistants/
FAQ (for Google + humans)
Is DLT only for school subjects?
No. DLT scores skills — not school content.
If a skill is real, it must be built (Depth), must hold under pressure (Load), and must generalise (Transfer).
Why does this matter to parents?
Because it ends the repeating cycle of:
“Try harder” → burnout → frustration → disengagement.
DLT tells you what to fix first.
Where should I start?
Start at the hub, then mount DLT:
https://edukatesg.com/education-os/
https://edukatesg.com/dlt/