yosui262

.................................................................

Super Extreme Engineering (SXE)

November 28, 2025

Overview

Super Extreme Engineering (SXE) is a framework for maximizing quantitative outcomes by updating team values and habits at high frequency.

SXE extends the philosophy of Extreme Programming to all human endeavors involving the construction of structures to achieve outcomes, reconstituted from an engineering perspective. While traditional Extreme Programming uses iterations of one to two weeks, SXE is characterized by observation and adjustment cycles of less than one day.

Model Structure

Components

SymbolNameDescription
xIndividual values and habitsValues, habits, skills, and information held by each worker
yTeam values and habitsJudgment criteria and priorities shared across the team
αTransformation function (x → y)Rules that convert individual values into team values. Designed and adjusted by humans
βTransformation function (y → z)Actual work output executed by the team. Depends on workers' information
zQuantitative outcomesObservable outcome metrics defined by the person who decided to adopt the framework

Transformation Chain

x  (α)  y  (β)  z

Humans input two parameters: α (transformation rules) and z (target outcomes).

Design Principles

Priority of y

Team values (y) always take priority over individual values (x). This structurally eliminates noise from information overload and delays in decision-making. Even when individuals are uncertain, clear y provides a single judgment criterion.

Individual expertise and on-site insights (x) are abstracted and reflected in α when sufficiently important. The priority of y does not mean x is ignored.

Nature of β

β (actual work output) depends on all dimensions of information held by workers. Information includes skills, knowledge, and accessibility to context. β is not a target for adjustment but for observation, functioning as an intermediate indicator to verify whether α adjustments are effective.

Managing others' capability development or skill improvement makes problems ambiguous and increases overall uncertainty. Therefore, direct adjustment of β is outside the scope of this framework.

Nature of z

Outcomes that are difficult to measure but important cannot be called outcomes. z must be quantitatively observable, and the z decision-maker updates it regularly.

Separation of Fixed and Variable

FixedVariable
Morning and evening meetings as rituals (framework)Content of y (team values)
Content of α (transformation rules)

This design philosophy minimizes what stays fixed and changes what is variable at high frequency.

Implementation

Concrete Implementation of α: Morning and Evening Meetings

α is implemented not as complex documents or tools, but as rituals. Morning and evening meetings are the fixed minimum habits.

RitualFunction
Morning meetingSynchronize today's y (team judgment criteria and priorities) across all members
Evening meetingEvaluate α based on observed z results and adjust α for the next day

Scope of Authority in Evening Meetings

Only α can be adjusted in evening meetings. The definition of z is under the authority of the z decision-maker and cannot be changed in evening meetings.

When gaps are detected, the focus at the operational level is on "how to change the approach (α)." If there is a problem with the definition of z itself, that recognition is raised as information to the z decision-maker and considered in weekly or longer cycles.

Benefits of Implementation as Rituals

By forcing everyone to be in the same place twice a day, the problem of communication costs is solved. Update costs are extremely low, interpretation discrepancies are immediately corrected in conversation, and high-frequency updates are automatically maintained as long as the ritual does not become mere formality.

Feedback Loop

The SXE feedback cycle is as follows:

Observe z  Detect gaps from expectations  Adjust α  y changes  Work appearing as β changes  z changes

Running this cycle in less than one day (preferably one cycle from morning to evening meeting) is the core of SXE.

z Observation and Update Cycle

z is observed and updated in the following three-layer structure:

CycleContent
DailyObserve z, adjust α
WeeklyObserve and evaluate z, review z definition if necessary
MonthlyObserve and evaluate z, review z definition for medium to long term

In the daily cycle, only α is subject to adjustment. Review of z definition is done in weekly or longer cycles.

Fractal Structure

Basic Unit

The basic unit of team size is three people. By nesting this unit in a fractal structure, it can adapt to any team size.

Connection Between Layers

When there are multiple layers, upper and lower layers are connected not by "propagation of y" but by "agreement on z definition."

The z decision-maker of the upper layer discusses with the z decision-maker of the lower layer and defines the z of the lower layer. This preserves the autonomy of α adjustment for each team. The upper layer does not intervene in the "approach" of the lower layer, only holding "what counts as an outcome."

Layered Rituals

Morning and evening meetings are held separately by each team. Synchronization between layers is done through weekly z sharing meetings by z decision-makers.

LayerRitual
Each teamMorning and evening meetings (less than daily)
Between z decision-makersz sharing meeting (weekly)

In z sharing meetings, information is shared only on the z axis for understanding the overall structure.

z Maker

Role

The z decision-maker has the following responsibilities:

  • Define z
  • Build consensus on z definition between layers
  • Judge the timing of z updates
  • Coordinate with the upper layer (if one exists)

Requirements

The following abilities and conditions are required of the z decision-maker:

  1. Deep understanding of SXE
  2. Strong commitments
  3. Ability to practice statistical and mathematical methods
  4. Ability to respond flexibly
  5. Ability to do engineering

Conditions for Success

The following conditions are necessary for SXE to function:

  1. z must be observable in real time. Outcome metrics must be measurable in less than one day.
  2. The cost of changing α must be sufficiently low. This is ensured by the rituals of morning and evening meetings.
  3. Adaptability of participants. Whether participants can adapt to high-frequency updates is mutually confirmed at the stage of joining the team.