Skip to main content

5th Grade Summary

Kam AI is not trying to win because it is the cheapest place to ask a sports question.

Kam AI is trying to win because it helps users reach better outcomes.

That outcome is not a guaranteed bet. It is a better decision loop.

The React Native app is where the user watches games, asks Kam, opens details, checks saved reads, and reviews decisions.

The backend is where the durable work happens. It owns routing, read models, source truth, answer contracts, and production validation.

The web app is now the public education layer. It should explain how Kam works and why the workspace helps users avoid stale decisions, save stronger theses, and learn from the result.

The purpose of Kam AI

The purpose of Kam AI is to give users an outcome edge through better process.

That means the product should help a user answer practical questions:

  • What changed since I last checked?
  • Is this source fresh enough to trust?
  • Did the market already price this in?
  • What is missing before I act?
  • Should I track, save, compare, review, wait, or pass?

The current Kam product is centered on the React Native app because those questions are workflow questions, not one-off chatbot questions.

The backend is not a helper behind a static page. It is the system that makes the workflow trustworthy by owning durable work and read truth.

Current Kam AI

The surfaces that create outcome edge

Home base

Summary Island

Shows what deserves attention before the user burns time checking everything.

Monitor

Watchlist

Keeps the user focused on the games, teams, markets, and futures that matter.

Explain

Ask Kam

Turns a question into a source-aware read with visible caveats.

Act

Platform Panel

Routes the answer into a concrete next step instead of leaving it as prose.

Inspect

Platform Detail

Lets the user inspect market context before trusting a read.

Receipts

Backend

Keeps source freshness, read models, evidence, and answer contracts behind the UI.

Takeaway: Kam AI should be explained as a coordinated outcome system backed by server-owned truth.

The real outcome loop

The app should not guess from raw data.

It should send compact, typed context to the backend and render the answer packet that comes back.

Workflow

React Native to backend to better decision

Kam should help the user move from a question to evidence, caveat, decision, result, and review.

  1. 1

    User opens a workspace surface

  2. 2

    React Native sends typed context

  3. 3

    Backend routes the turn

  4. 4

    Read models load exact-key truth

  5. 5

    Answer contract and receipt are built

  6. 6

    React Native renders chat, panel, summary, or detail

  7. 7

    User tracks, saves, compares, reviews, waits, or passes

The moat is not one screen. It is the loop that turns source truth into a reviewable user decision.

What each side should own

Current ownership boundary

Layer
React Native
Owns
visible state, handoffs, layout, workspace actions
Should not own
durable source truth
Layer
Backend
Owns
routing, tool policy, read models, answer contracts
Should not own
local UI layout
Layer
Materializers
Owns
current and historical read models
Should not own
user-facing prose
Layer
Web
Owns
public explanation, docs, blog
Should not own
primary workspace behavior

Takeaway: Good architecture matters because it protects the outcome: no local guessing, no hidden stale context, and no vague answer without a next step.

Why users should care

This outcome language matters because sports research is messy.

A user may ask about a line move, a saved read, a prediction-market signal, a watchlist spot, a player prop, a futures market, or a stale board.

If the app guesses locally, it can sound confident while missing the truth path.

If the backend owns the read path, Kam can show source freshness, missing context, caveats, and next actions in the same workflow.

That is how Kam competes on outcomes: fewer rushed decisions, fewer stale reads, stronger saved theses, and a better review loop after the result.

Next action

The public site should now point users toward two ideas:

  • how the React Native workspace creates a better decision loop
  • how backend read truth becomes source-aware reads, saved theses, and outcome review

Everything else should be rebuilt from current architecture, not preserved as legacy blog inventory.

Research question

How does Kam AI create an outcome edge?

Kam AI creates an outcome edge by combining a React Native decision workspace with backend-owned source truth, freshness receipts, saved reads, and review loops.

Current product map

What creates the Kam outcome edge

Decision loop

React Native

Starts work, renders state, opens the right surface.

Source truth

Backend

Owns durable runs, routing, tool policy, answer contracts.

Fresh reads

Materializers

Prepare current and historical read models before chat asks.

Education

Web

Explains how the product improves outcomes.

Final call

User

Keeps the final judgment human-controlled.

Private

Raw traces

Receipts are public; debug payloads are not.

Takeaway: The current Kam AI moat is coordination across workspace surfaces, backend-owned truth, and reviewable outcomes.

React Native workspace surfaces

Surface
Summary Island
Job
Home base, tools, memory, market shortcuts
User outcome
See what needs attention
Surface
Watchlist
Job
Saved games and monitoring
User outcome
Keep tracked spots in one place
Surface
Ask Kam
Job
Chat, stream, trace, task context
User outcome
Ask what changed and why
Surface
Platform Panel
Job
Typed widgets and handbacks
User outcome
Turn answers into workspace actions
Surface
Platform Detail
Job
Game-level market context
User outcome
Inspect sources before acting

Takeaway: The surfaces should feel like one outcome loop, not a set of disconnected marketing pages.

Current Kam AI outcome flow

  1. 1React Native workspace
  2. 2Typed context
  3. 3Backend route and turn contract
  4. 4Materialized read models
  5. 5Product answer packet
  6. 6Chat, panel, summary, detail
  7. 7Save, track, compare, review, or pass

Takeaway: React Native starts and renders the work; the backend owns the durable run and source truth that make review possible.

Read next

Related field notes

View all posts
backend-contractsturn-envelope

How Backend Data Becomes a Better Outcome

See how Kam turns backend receipts into source-aware reads that help users track, save, compare, review, or pass with more discipline.

8 min read