Study Hub
Menu
Filters
11 Study ItemsCanonical content from docs/study

now

  1. 01Orient to the calculation-path coachDefine the active Standard Trainer direction, its learner outcome, and the authority boundary between deterministic tools and AI assistance.90 min · not started · AI Product Management
  2. 02Frame the learner failure with evidenceTranslate calculation-process observations into a testable problem statement and interview plan.105 min · not started · AI Product Management
  3. 03Model solution graphs and student stepsRepresent the canonical Kp calculation path, structured student work, example tracing, and model tracing as separate inspectable objects.120 min · not started · Calculation Trainer
  4. 04Freeze authority, policy, and provenanceBind a curated problem, solution graph, marking policy, and review state before making correctness or ECF claims.100 min · not started · Calculation Trainer
  5. 05Design deterministic validatorsSpecify reproducible validators for arithmetic, units, expressions, species, dependencies, significant figures, and mark allocation.120 min · not started · Agent Engineering
  6. 06Localise the first invalid step and bound ECFAlign structured work to the solution graph, identify the earliest invalid step, classify misconceptions, and apply bounded error-carried-forward rules.135 min · not started · Calculation Trainer

next

  1. 01Route tools with inspectable tracesDesign constrained runtime routing across validators, diagnosis, marking, and hint paths without granting the model authority it does not own.110 min · not started · Agent Engineering
  2. 02Plan hints and reviewed follow-up practiceTurn bounded diagnoses into hint decisions, learner weakness updates, and reviewed next-problem selection.120 min · not started · Calculation Trainer
  3. 03Evaluate components and trajectoriesConsolidate the vertical slice with component-level correctness checks and end-to-end trajectory evaluation before claiming readiness.120 min · not started · Product Delivery and Evals

later

  1. 01Run bounded learner evidence sessionsObserve real use, separate learning evidence from product and review evidence, and permit only a bounded evidence-backed change.150 min · not started · Product Delivery and Evals
  2. 02Turn delivery evidence into interview storiesTranslate the calculation-trainer work into concise role-relevant stories, portfolio proof, and application evidence without overstating claims.135 min · not started · Interview Preparation