what we build — opening
Opening
Volume II ended with a brief on the table — agreed on by the trio, signed, the baseline measured. Gal still grades exams the way she always has. Uri still runs his Cardcom reconciliation every month. The understanding is intact. Now it crosses its first real boundary.
This volume is what happens to the brief between the moment it is signed and the moment a developer pulls a story into In progress. The Feature Brief contains a witnessed understanding of a problem. That understanding is now going to be translated into the artifacts a team builds with: Epics in JIRA, story maps in Miro, wireframes in Figma, schemas in the database, API contracts between services, scenarios in test files. Each of these is a different shape. Each can carry meaning forward — or quietly replace it with its own shape.
What gets built is downstream of which shape catches the meaning. An Epic named "Teacher Dashboard Improvements" and an Epic named "Grade exam results and publish to the school system" point a team toward two different products. A schema with a charge_attempts table and a schema without it produces two different applications. The choices are not neutral. The shape of the holder shapes what gets held.
The volume in one sentence. Each phase carries the brief's meaning at a higher resolution than the one before. A phase is done when the resolution is enough — not maximal, not minimal — to clear the next gate without inventing things downstream.
The shape of this volume
Seven phases. Each one increases the resolution of understanding by one step. Each one has a gate — a bar of enough that the work has to clear before the next phase can begin. The gates are not ceremony; they are the test of whether meaning is still travelling. When a gate is treated as a checklist instead of a judgment, the meaning has already been replaced.
- The Translation — what happens at the boundary, why the unit of work itself is a meaning-carrier, the failure modes named.
- Epics — naming activities, kicking them off, drafting story backbones. Technical Epics where the witness is a system. Resolution: enough to map.
- Design Shaping — flows, states, wireframes for the current release. Just-in-time, not upfront. Resolution: enough to see the moments.
- Technical Shaping — sequence diagrams, schemas, API contracts, ADRs, ilities. Same just-in-time discipline. Resolution: enough to write Givens that mean something specific.
- Planning — the convergence point of the volume. Everything before it is preparation; everything after it executes what it decides. The story map, the walking skeleton, INVEST as a story-shape diagnostic, what kanban replaces. Resolution: enough to know what flows next, and why.
- Amigos — three-person session per story. Gherkin from the person's situation. Scenario names. Resolution: enough to pull.
- Closing — the bridge to Volume IV.
In the operational framework, this volume describes the Scope phase — the work between the witnessed problem (Discovery, in Volume II) and the shipped change (Execution, in Volume IV). The Prediction lives across all of them: it was named in Volume II as a baseline measurement, it must travel through Scope intact in this volume, and it will be checked in Volume V. Volume III holds Scope alone — the translation of a signed brief into a story ready to pull.
A note on vocabulary: this volume describes a lean / kanban execution practice, not scrum. Work flows; it is not packed into sprints. The team pulls stories under WIP limits rather than committing to a sprint backlog. Iterations are framed as releases — what ships at the end of a slice. Daily and retrospective conversations are kept, but they are about flow, blockers, and pull, not about velocity, points, or sprint goals.