Skip to Content
image description
image description

Sample spec

Sample Discovery Sprint output

The actual spec we put on the contract. See it before you book.

A redacted, real Discovery Sprint deliverable from a production manufacturing scheduler rebuild. Five sections, 24 pages, written in plain English by senior engineers — the same document your fixed-price contract is built on.

Read it before the call. If the depth and specificity match what you’d expect on your own modernization, the Discovery Sprint is the right next step.

  • 01 Scope. What’s in the rebuild and what’s explicitly out, with rationale for each cut.
  • 02 User stories. Numbered acceptance criteria the AI coding agent executes against and the team verifies against.
  • 03 Risks. Ranked technical and operational risks with mitigation plans and trip-wire conditions.
  • 04 Architecture. Stack decision, data model migration plan, integration boundaries.
  • 05 Implementation plan. Week-by-week sequencing, milestones, and the production-readiness checklist.
01 · What's inside

The five sections, every time

Same structure on every Discovery Sprint. The depth varies with the project; the structure does not.

SECTION 01

Scope

What's in, what's out, what migrates as-is. Exclusions are explicit, with the reasoning written down.

SECTION 02

User stories

Numbered (US-01, US-02…). Each has acceptance criteria the AI agent executes against.

SECTION 03

Risks

Ranked R-01 through R-N. Severity, owner, mitigation, and trip-wire condition for each.

SECTION 04

Architecture

Stack, data model migration, integration boundaries. Decisions stated; alternatives ruled out.

SECTION 05

Implementation plan

Week-by-week sequencing. Production-readiness checklist. The contract is built on this.

02 · Page sample

One page from a real spec

Excerpts from sections 02 (user stories) and 03 (risks) of a manufacturing scheduler MES rebuild. Names redacted; structure intact.

DISCOVERY SPRINT · ACME-MES-2026 Confidential · NDA
SECTION 02 · USER STORIES

Production scheduler — core flows

14 stories total in this section; two sample stories shown below. Acceptance criteria are the contract between the team and the AI coding agent.

US-12 · MUST · WEEK 4
Schedule a production run from a confirmed sales order

As a production planner, I want to schedule a production run directly from a confirmed sales order so I do not re-key part numbers and quantities into the scheduler.

Acceptance: Given a sales order in status “confirmed,” when I click “Schedule run,” then the scheduler creates a draft run pre-populated with order line items, customer, requested ship date, and estimated cycle time pulled from the routing master. The draft run cannot be saved until a work-center assignment exists for every line item.

US-13 · MUST · WEEK 4
Detect and surface scheduling conflicts before run commit

As a production planner, I want the system to detect scheduling conflicts (work-center overbooking, material-availability gap, operator unavailability) before I commit a run.

Acceptance: When I submit a draft run, the system runs three conflict checks (capacity, material, labor) and either commits the run or returns a structured conflict report listing each conflict with the conflicting run ID, the resource at issue, and the time window. No run can commit with an unresolved conflict.

SECTION 03 · RISKS

Top-ranked risks

ID Risk Severity Mitigation
R-04 Routing master data quality is unverified; bad data will silently corrupt new schedules High Audit routing master in week 1; flag missing cycle times before US-12 build starts.
R-07 Single PHP 5.6 ETL job feeds the legacy system; rewrite is in scope, but cutover order is fragile High ETL rebuild scoped first (week 2). Dual-write window for two weeks. Rollback to legacy ETL retained until week 10.
R-09 Operations team built a manual workaround for capacity check; workaround logic is undocumented Medium Capture workaround logic in a 2-hour shadowing session, week 1. US-13 acceptance covers the documented logic.
R-12 Single sponsor (VP Operations) is the only weekly decision-maker; unavailability stalls the project Medium Backup sponsor (Director Operations) named in contract. Decision authority documented at sprint kickoff.
Page 8 of 24 · DISCOVERY-SPRINT-ACME-MES-2026.pdf
03 · Why look at this before booking

The spec is the product. The Discovery Sprint produces the spec.

Three reasons to read this PDF before you schedule the call.

01

You see the depth before you sign anything

The 90-day fixed-price contract is built on a document like this one. Read it; decide if the depth matches what your modernization deserves.

02

It signals what's in scope and what isn't

Discovery Sprints walk away from projects we can’t ship in 90 days. The “Scope” section is honest about exclusions; the “Risks” section is honest about uncertainty.

03

You arrive at the call calibrated

Most prospects ask the same five questions on call one. Reading the spec sample first means call one is about your modernization, not the methodology.

20+

Years shipping
production software

400+

Client problems
solved

90

Days, fixed
price, every time

25%

Discount if we
miss the deadline

image description
image description

Read the spec. Then talk to an AI engineer.

If the spec sample matches the depth your modernization needs, the next step is a 30-minute call to confirm fit.

Back to top