AI planning + lifecycle cost modeling · Aloha AI
AI Build Cost Workbench
Live interactive cost-modeling workbench · v2 released August 2026
Turn an early AI-build idea into a transparent, reviewable scenario spanning implementation labor, model usage, infrastructure, vendors, human operations, uncertainty, and first-year cost.
Controlled visual demonstration
Current live interface
This sandboxed preview shows the current public interface. Open the full site to use the build in its intended window.
What I noticed
Custom AI projects are often discussed as a single price even though discovery, research, architecture, implementation, testing, documentation, launch, and maintenance require different kinds and amounts of labor. That opacity makes early scoping difficult for both a buyer and a builder.
What I researched
The v2 method draws on cost-estimation practices that emphasize a defined baseline, documented assumptions, ranges, sensitivity, uncertainty, and lifecycle cost. It also incorporates AI-lifecycle review lanes and a dated, editable model-price catalog rather than hiding a universal project-price claim behind a project-type selector.
What exists
A six-stage browser workbench covering project brief, editable labor architecture, model/API workloads, infrastructure, one-time and recurring costs, named uncertainty drivers, review lanes, specification confidence, cost composition, sensitivity drivers, build and monthly ranges, and first-year cost. Scenarios can be saved locally, shared by URL, printed to PDF, or exported as CSV and JSON.
Who it serves
Founders, creators, independent builders, consultants, prospective buyers, and teams preparing an AI-build discovery or procurement conversation.
What it demonstrates
Product strategy, lifecycle cost modeling, AI workload economics, transparent assumptions, uncertainty analysis, interaction design, local-first privacy, export architecture, and evidence-linked methodology.
Current evidence
The canonical v2 production release was verified end to end: example loading, live recalculation, eight-workstream labor modeling, multiple model workloads, persistence across reload, share-link reconstruction, CSV and JSON generation, result composition, methodology/privacy/terms routes, and production security headers. The legacy lead relay remains disabled. Outputs are user-authored planning scenarios rather than quotes or independently validated market benchmarks.
Verified build record
What is actually running, and what it is not.
LiveLive · v2, August 2026
The problem. Custom AI projects get quoted as one number, even though discovery, research, architecture, implementation, testing, documentation, launch and maintenance are different kinds and amounts of work. That opacity hurts the buyer and the builder equally.
What it does. Walks six stages — project brief, editable labour architecture across eight workstreams, model and API workloads, infrastructure, one-time and recurring costs, named uncertainty drivers — and returns cost composition, sensitivity drivers, build and monthly ranges, and a first-year total. Scenarios save locally, share by URL, print to PDF, and export as CSV and JSON. The model-price catalogue is dated and editable rather than hidden behind a project-type selector.
Built on. A static browser application with local persistence and URL-encoded scenario sharing, served from Vercel.
What it does not establish. Outputs are the user's own planning scenarios. They are not quotes, and they are not independently validated market benchmarks.
This record is generated from the same inventory as the complete build index: the production URL was checked at the last regeneration, and the status line above repeats what the build reports about itself rather than restating an intention.