Agentic LE Capability Matters
Earlier fixed-workflow studio

Flower experiment studio

About this resource · How to use it · Learning engineering connections

What is this resource?

The earlier fixed-workflow experiment studio, preserved for its configurations and original records. New work on dynamic conversations and two-phase deployment tests uses the separate revised tool.

How should I use it?

  1. Choose a fictional scenario and the full flower team. Inspect domain assignments, overlapping responsibilities, behavioral settings, conditions, and fixed challenges.
  2. In Review & run, choose Flower: five conditions and the domain and member to compare. Inspect call counts: domain omissions preserve contribution opportunities, while member removal changes capacity. Export configuration or run through the local runner.
  3. Replay a saved run to inspect prompts, evidence, and artifact revisions. The original declared-design synthetic outcome cards remain separate from the new blind evidence scorecard.
  4. In Scorecard, prepare opaque packets and keep the private assignment key separate. Use the isolated blind workspace to cite evidence, retain unknowns, and freeze all reviews before revealing assignments or comparing candidates within a scenario.
  5. Fork an iteration when changing a team, assumption, or design. Export the preserved snapshot, frozen scorecards, and original journals to retain the complete parent-linked evidence trail.

How does this connect to LENS and learning engineering?

The flower organizes complementary contributions around shared understanding. Learner and stakeholder responsibilities remain shared; represented expertise does not establish proficiency. LENS retains Human flourishing, Learning sciences, and Systems engineering as its foundations. The evidence scorecard separately examines human flourishing support, system performance, and learning; simulated outputs do not establish outcomes with people.

This studio runs the earlier fixed schedule. For dynamic conversations, bounded expertise, and easy/medium/hard deployment tests, open the revised two-phase tool.

Configure eight starting expertise domains around shared understanding. Explore how responsibilities, overlap, team capacity, and operating conditions shape a design.

Checking runner…

A shared center and eight starting domains

Shared understanding and responsibility to learners and other stakeholders apply to every member. The flower organizes expertise; it does not prescribe a single team or establish proficiency.

Explore the flower and its source · Read the five comparison conditions

The default team uses 40 calls. Global omission and one-member withdrawal retain those opportunities; removing one default domain member leaves 36. A single agent uses the full reference schedule. Review the resolved plan when changing assignments or stages.

Loading the experiment catalog…

Preparing the configuration workspace…