Loading…
Loading…
EDGE Deploy stage
Explore produces a signed BRD. The AI Delivery Chain turns that BRD into a Solution Design, Proposal, scripts, and runbook inside Deploy. Govern locks the gates. Enable cuts over. See the full methodology →
Inside one chain run
Every downstream stage gets the upstream artifact pinned verbatim. Conflicts are flagged in a Notes paragraph.
AI agent captures requirements across web, WhatsApp, SMS, or email. Structured fields: entities, users, modules, integrations, data sources, timeline, constraints.
Stage 1 of the chain: Business Requirements Document with functional requirements, integration map, data migration scope, and a risk + assumptions register.
Stage 2: technical architecture, module design, integration design, data model, security and access control. Generated with the BRD pinned verbatim as upstream context.
Stage 3: scope, methodology, deliverables, milestones, team composition, and a transparent pricing band. BRD + Solution Design pinned upstream so nothing drifts.
Skill-driven file expansion produces per-file deliverables (NetSuite scripts, configs). The deterministic Value Analysis quantifies hours and dollars of authoring effort eliminated.
How skills steer the chain
Consultant-authored skills override outline, conjure deployment files, and anchor the model to curated references.
An instruction-role skill can declare its own outline. replacing the chain's default section list with one tuned to a specific industry, platform, or contract template.
A `manifest:scripts` directive prompts the AI to enumerate every deployment file the engagement needs at runtime. Each file becomes its own per-file generation iteration.
`file:<path>` and `script:<name>` directives produce separate `type='script'` deliverables (NetSuite SuiteScripts, configurations, integration templates) alongside the prose artifacts.
Skills can attach reference URLs. When grounding is on, the chain fetches each URL through a curated proxy and inlines the content as preferred source material.
Transparency
We're explicit about the AI's role. Every AI output is reviewed by a consultant before it reaches you.
Policy: No AI-generated BRD or proposal is sent to a client without explicit senior consultant review and sign-off. Always.
What you get
Every chain run produces all six artifacts. Each is versioned, AM-reviewable, and the chain is re-runnable end-to-end.
Stage 1 of the chain. Functional requirements, integration map, data migration scope, risk + assumptions register, and an estimate range.
Stage 2. Technical architecture, module design, integration design, data model, security & access. Synchronized with the BRD.
Stage 3. Scope, methodology, deliverables, milestones, team composition, and a transparent pricing band. Grounded in BRD + Solution Design.
Per-file `type='script'` deliverables. NetSuite SuiteScripts, configurations, integration templates. produced by skill-driven manifest expansion.
Deterministic markdown deliverable quantifying hours and dollars of authoring effort eliminated. No LLM call at generation time. Re-runnable, reproducible.
Every chain run is versioned, every artifact is logged, and the source skills are recorded. Buyer, AM, and consultants see the same provenance.