Naru AI

Applied AI product company

Where new AIbecomes useful product.

Naru AI explores how emerging AI capabilities translate into useful real-world applications — from product design and system architecture through rapid development, launch, and market testing.

Scroll — 10 findings from building them

02 — Productsthe evidence
behind the findings

Where the findings came from.

Three products across three domains, each with a constraint we would not relax.

01

StemBuddy

education2026

Science investigations kids run themselves.

Science investigations for kids 8–12 — engineered so the AI cannot do the thinking for them.

02

Naru AI

health2025–26

A family health journal for iOS.

Family health journal — shipped through App Review with a paid tier, retired after market testing.

03

SEAM

fintech2026

A wallet for people banks leave out.

Mission-locked digital wallet and community fund, in active exploration — preview published.

03 — Approach

How we build.

01

Wired ≠ live

Nothing counts as shipped until it writes its first real production row. A feature that compiles, deploys and demos is still a claim, not a fact — which is why the StemBuddy lane above stops at 04 until real usage exists to read.

02

Policy compiles

Commitments live in CI, not in review discipline. On StemBuddy, a third-party tracker in the kid-facing bundle fails the build; so do leaked secrets and accessibility regressions. A promise that can’t fail a build is a hope.

03

One writer, many agents

Builds run under a designed multi-agent coordination protocol: a single-writer state snapshot, an append-only broadcast log, and mandatory human gate-reviews on billing, migrations and anything a child can type into. 1,600+ commits since January under that system.

04 — Contact

Tell us what you're building.

We take on work where getting it wrong has consequences — products for children, health records, money. If that is the shape of your problem, we should talk.

  • New buildAn AI product to take from idea through launch and market testing.
  • Adding AIAn existing product that needs AI in it without eroding the trust it already has.
  • Second opinionAn AI product already underway, and a read on whether it will hold up in front of real users.

min@naru-ai.com

Product repositories are private: two touch children's data and one sits beside real bank transaction data. Happy to walk through any of them on a call.

Contact

Tell us what you're building.

We take on work where getting it wrong has consequences — products for children, health records, money.

No form, no tracker, nothing loaded from anywhere else — just an address.