Sequence learningINTERACTIVE
Follow the pulse.
Watch the illuminated sequence, then repeat it. Each successful round adds one step.
Awaiting session
Software demo
Explore the Hap Pines software experience.
This interactive demo presents selected workflows from the Hap Pines platform, including structured cognitive tasks, interaction telemetry, adaptive task logic and AI-generated insight views. It is designed to show how the complete software experience is being built and how its core systems work together.
Public software demo: this page is an interactive demonstration of selected Hap Pines product workflows. It does not create real user accounts or run the complete live Hap Pines model pipeline. The full software is under active development and has not yet been publicly released; example outputs shown here are illustrative.
HAP / AIAdaptive signal engine
TASKTIMEERRORSHIFTPACECUE01 / Cognitive session
A task changes. The model watches what changes with it.
The platform varies the cognitive environment, captures response behavior and compares repeated interactions so that insights are grounded in patterns across conditions.
▲ ● ▲ ● ▲ ?
02 / Micro-lab
Three different mechanics. Three different kinds of signal.
These interactive modules show how different task mechanics can expose memory, inhibition, sequence learning and recovery behavior as structured inputs to the analysis pipeline.
Working memoryINTERACTIVE
Hold the pattern.
Four cells flash briefly. Re-select them after the pattern disappears.
Awaiting session
Interference controlINTERACTIVE
Name the ink, not the word.
A conflicting word competes with its display colour. Choose the display colour.
GREEN
Awaiting response
These demo modules illustrate the task and telemetry layer of the Hap Pines software. Full profile generation is part of the complete application and model pipeline currently being developed; this public demo focuses on the interaction workflow.
02 / Adaptive engine
The adaptive engine changes the task environment.
It changes difficulty, cues, timing and task rules so the model can compare performance across controlled conditions.
Condition AVisual cue firstShape hierarchy visible
Condition BText instruction firstNo visual preview
Condition CRule switchPrevious cue becomes unreliable
Condition DTime compressionSame task, shorter window
Model question
What moved when the environment moved?
Accuracy+12%illustrative
Response time−0.8sillustrative
Error recoverystrongerillustrative
The system tests whether conditions such as visual-first presentation repeatedly correlate with stronger performance and measures how consistently that relationship holds.
03 / Pattern synthesis
From raw behavior to a readable profile.
The model converts internal signals into human-readable insights that show the recurring pattern, the supporting evidence and the level of confidence behind it.
SESSION SIGNALSWEIGHT
Visual-first accuracy delta.82
Rule-switch recovery.76
Long-session variance.71
Immediate-feedback recovery.67
Single-task result.19
→
ILLUSTRATIVE OBSERVATION
Performance remained more stable when visual structure preceded written explanation.
The evidence strengthens when the same relationship repeats across task types and sessions, while the platform keeps the interpretation tied to observed conditions.
Evidence confidence
developing
04 / Session replay
See where the behavior changed.
Profiles retain traceability to the sessions and condition changes that contributed to each insight, giving users a clear evidence path behind the model output.
Accuracy trace / illustrative
ConditionBaseline
Model noteStable reference point
Accuracy92%
Latency1.8s
04 / One profile, different explanations
Present the same evidence for different users.
The underlying evidence stays consistent while the interface presents the level of detail and action appropriate for each authorized audience.
YOUR CURRENT PATTERN MAP
14 sessions · illustrativeWhat seems to help you perform better.
Visual structure before long instructions
Your accuracy and pace were more stable when a diagram, pattern or example appeared before a longer written explanation.
Recovery after rule changes improved across recent sessions.
Stability currently peaks in shorter focused blocks.
05 / Strategy engine
Turn supported patterns into practical strategies.
The strategy layer converts supported patterns into practical experiments users can test across learning and performance settings, with results feeding back into the profile over time.
LEARNING STRUCTURE
TESTExample before abstraction
Introduce one worked example before the formal rule, then compare retention against the usual order.
SESSION DESIGN
TESTShorter focused blocks
Split a long session into smaller blocks and observe whether accuracy remains more stable.
FEEDBACK TIMING
TESTImmediate correction loop
Provide feedback directly after an error and compare recovery on the next related task.
Software development status
The complete Hap Pines application is under active development.
The full Hap Pines software has not yet been publicly released. This demo gives visitors access to selected product interactions while the complete application, account system, model pipeline and profile infrastructure are being completed for release.