The Anatomy of an A.I. Product

1. Whats AI?

1. "Artificial Intelligence" is a bad name 2. It is an intelligence amplifier - so were: the printing press, the telegragh, the search engine - makes plans and outputs abundunt 3. Like nothing else before, its programmatic decision making - this is profound - aka "conditional branching on MEANING"

2. The Creator(s) #### Applies to teams or solo builders

Must haves: - access to coding agents - a strong vision - comfort w/ ambiguity - willingness to iterate & throw work away

Good to have: - compelling taste - UI and UX expertise - software development expertise

Must Orchestrate: - strategy agents - coding agents - design agents - brand identity - user communications (marketing, support)

3. The A.I. Product #### A Ratchet of Cumulative LLM Outputs

Whats an "A.I. Product"? - a.k.a. "AI-Native Product" - A product that holds the profound invention as central to its operation and existence - A product whose core value would be infeasible without learned inference - A product for which intelligence is the operating principle, not a feature

Users use A.I. products to: - decrease cost - save time - improve outputs

Attributes: - makes choice cumulative - couples the builder's differentiated knowledge w/ a ratchet that maeks user's judgement cumulative - improves future work from past work learnings, with human oversight / management - canonical knowledge, from the creator, shapes the field of possibilities

UX Tenets: - Deliberate action outranks inference - Memory has scope - Promotion requires consent - Human attention goes to consequential choices - Let users accept, export and stop whenever the result succeeds - Valuable judgements are scoped & reversible

4. The Journey

1. The job is to build leverage on human judgement 2. The first leverage is the creator's differentiate knowledge - Knowledge sources - e.g. Wikipedia articles, YouTube transcripts, niche websites, scanned books - Knowledge graphs - a simple Postgres node/edge database , curated by the creator(s), makes knowledge lookups exponentially more performant / cheap - for LitScenes, a meaning graph with an interface that allows the apps to drive story and aesthetic from user provided media / chat - Compiled knowledge - e.g. "statements made at Hawaii community meetings in 2026 , extracted from the county's YouTube channel" - out-competes any language model on the subject of community concerns in Hawaii - LitScenes example: a visual style library 3. A loop develops:

i. Agents expand the search

• acquire sources
• * extract candidate knowledge
• * propose assets, workflows and implementations

ii. The creator judges what deserves development

• rejects, edits, combines and redirects
• * curates candidate knowledge and assets
• * selects workflows and implementations to test

iii. Agents make the judgments executable

• assemble knowledge and assets into context
• * translate intent into prompts, tools, schemas and interfaces
• * produce runnable variants in workbenches

iv. The creator puts the variants under pressure

• acts as user zero
• * compares finished results
• * evaluates quality, coherence, reliability and cost
• * discovers where the product’s assumptions fail

v. The creator promotes what the evidence supports

• canonical knowledge enters graphs and reference libraries
• * dependable workflows become product behavior
• * successful decisions become defaults
• * failures become eval and regression cases
• * friction becomes specifications for the next implementation

vi. The next search begins from more informed ground

• agents inherit the accumulated knowledge, assets, specifications and tests
• * settled decisions narrow the search
• * unresolved failures and missing knowledge reopen it
• * the loop begins again

The build ratchet does not preserve every output. It preserves the judgment earned by producing and testing those outputs.

5. Useful outputs

The output is the rendered video file — and, just as much, the media that composes it. The intermediate assets a render produces — a restyled still, a narration take, an isolated clip — aren't scaffolding you throw away once the final cut exists. They're outputs in their own right, useful the moment they're made.

What's next

Post two goes into inputs and outputs directly — how knowledge actually flows between the interfaces. In practice that flow starts with coding agents and a single-admin local workbench, then moves out to a mobile, desktop, or web app once it's proven.

One thing I'd flag now, before the deep dives: there wasn't a serious wrong turn building this. What saved me was staying extremely versatile. On earlier projects I just assumed web apps were the only real option — I don't know why, but that was the ceiling in my head. Fable 5 is what opened up desktop and mobile app feasibility for any project, anywhere. That's the mindset this whole series is arguing for: don't let the interface you already know how to build be the one you default to.

---

Want the live version of this? Join the Orchestrating the A.I. Product workshop at Entrepreneurs Sandbox in Honolulu, or explore the case study yourself at litscenes.ai.