Open-Weights Agent

A Next Right Thing proof of concept · Kai builds, Lucille manages, Roper sponsors · begun July 22, 2026

Can we stand up an agent on a Chinese open-weight AI model — with a front end Roper can message from anywhere in the world?

That is the whole project. If Roper can ping it from Kansas City and get a good answer back, it succeeded.

Now: Step 1 — Kai is scoping the hardware. Next action: Kai researches buy-vs-rent options and writes the short report. Lucille sets the first weekly check-in.

Why this project

The strongest open-weight AI models in the world right now come from Chinese labs — DeepSeek, Qwen, Kimi, GLM. Most US companies cannot or will not touch them.

Kai lives inside the Chinese firewall. He has native access to these models, their documentation, and local pricing. For this project, that is an advantage almost nobody else has.

The three steps

Step 1 — Scope the hardware ← now

Kai answers one question: what machine do we need — buy or rent?

Option A: buy a machine (for example a Mac or mini-PC with 32GB+ memory).
Option B: rent hosted compute (a cloud server or GPU, in China or Hong Kong).
For each option: cost, what size model it can run, and how hard it is to set up.

Deliverable: a short report with prices and Kai's recommendation, in about two weeks. Roper reads the report and approves the hardware before any purchase.

Step 2 — Stand up the agent

Get one Chinese open-weight model running on the chosen hardware (Kai picks the model). Make it an agent: it answers questions and can do at least one useful task.

Step 3 — Put a front end on it

Roper can message it from anywhere in the world — phone or laptop — and get answers. Kai picks the simplest path that works across the firewall. Deliverable: Roper pings it from the US, it answers, and Kai writes a short summary — what it cost, what he learned, what he would do next.

No deadline pressure — but if any step is stuck for more than two weeks, we pause and regroup instead of grinding.

Roles

Person	Owns
Kai — builder	All technical choices: hardware recommendation, model, setup, front end.
Lucille — manager	The plan, check-ins at least twice a week, keeping this page current, tracking spend and Kai's hours, raising blockers to Roper.
Roper — sponsor	Direction, funding, the hardware go/no-go after Step 1.

Hard rules

Chinese law comes first. Private project for us only — no public service, no bot open to users in China. If anything needs registration or ID verification Kai is unsure about, or feels legally risky in any way, he stops and flags it. Kai is never asked to take legal risk.
Absolute data wall. No NFP data, no client data, no personal data — ever. Test data only. Assume anything sent to a hosted Chinese API may be kept.
Budget. Running costs capped at $250/month. A hardware purchase is separate — one-time, approved by Roper after the Step 1 report. Lucille tracks all spend.
Security basics. Private repo only; no passwords or API keys ever committed; keys live in a password manager or a local .env file.
Ownership. Everything built on paid NRT time belongs to NRT.
Document in simple English as we go.

Decision log

Date	Decision
Jul 22, 2026	Project agreed. Scope shared with Lucille and Kai; Lucille confirmed the plan (via Narra restatement) and accepted the manager role. Budget cap set at $250/month. Check-in rhythm: at least twice a week.

Reports

Report 1 — hardware scoping. Coming after Step 1.