CL-00 STATUS: SHARED LEAD: HONGLENG TYPE: CAPABILITY
Shared AI / ML Capability

The shared AI layer.

A reusable AI/ML capability that products can adopt when the use case and safety boundary fit. Kasekor inside Chomkar is the current verified implementation.

Back to work See it in Chomkar →
WHY IT IS THE MOAT

A capability, not a feature.

Most teams rebuild the same AI plumbing for every product. We refuse to.

/ 01

Build it once

Prompts, voice, routing, guardrails, and evals are reusable building blocks. We verify one narrow capability first, then make it available where it genuinely fits.

/ 02

Reuse everywhere

A product may reuse a verified component instead of starting from zero. Products without a justified AI need remain independent of the layer.

/ 03

That compounding is the moat

The advantage is disciplined reuse: each proven component can lower later delivery cost without turning a roadmap assumption into a shipped claim.

LIVE TODAY

Live today: Kasekor.

The first thing running on the layer is already in a real product's hands.

LAYER v0 — KASEKOR STACK: Claude Haiku + ElevenLabs Khmer voice + Cloudflare AI Gateway CONSUMERS: Chomkar NEXT: SAT Digital

Kasekor is a Khmer lesson-helper living inside Chomkar — it explains farming concepts to growers in their own language. It is firewalled from anything commercial: it never touches price, demand, or quantity. Teaching is its only job, and that boundary is deliberate.

HOW PRODUCTS USE IT

A service with a roadmap.

The layer is run like a product, not a help-desk. Hongleng is its lead — with the right to say "not this week."

01

Request

A product team brings a real need — a Khmer explainer, a classifier, a voice flow — and writes down what success looks like.

Owner: product team
02

Prioritise across products

Requests are weighed against everything else the layer serves. Some ship now; some wait. "Not this week" is a real and frequent answer.

Owner: Hongleng
03

Reusable component

What gets built is designed to serve more than one product — a shared prompt, model route, or guardrail — not a one-off bolted onto a single app.

Owner: Hongleng
04

Ship

The component lands in the layer, the requesting product consumes it, and the next product inherits it for free. The roadmap moves on.

Owner: shared layer
LEAD

Who runs the layer.

One owner, accountable for the reusable capability and every approved adoption.

03AI / ML LEAD
Hongleng

Hongleng

Computer Science · leads the shared AI/ML layer

Leads the reusable AI capability, verifies its boundaries, and decides what is ready for another product to adopt.

Credited on: the shared AI layer · Kasekor
NEXT

See where it runs.

The layer is only as real as the products on it. Start with the one it lives in today.

See it in Chomkar → Inside the Lab