AnythingChina: connecting logistics experience with AI-assisted sourcing
AnythingChina is my attempt to connect experience in logistics and factories with AI-assisted sourcing. AI can research products, translate requirements and organize supplier leads. Whether a factory still operates at an address, who can respond to a request and how to obtain a sample still require contact with people.
I am He Fangsheng, also known as Hequbing. My earlier work included courier communication tools, advertising and business customers. That experience led to a practical question: how can existing local connections help a customer complete a specific task?
A search result is an initial lead
Suppose a buyer needs a factory willing to produce a small customized batch. An online list can help narrow the search, but product scope, delivery terms, samples, contacts and site conditions still need confirmation. A photograph seen by AI is not a site visit; an old web page does not establish current production capacity.
AnythingChina aims to bridge that gap: turn a company name into a reachable contact, send requirements to someone willing to respond, and arrange authorized local follow-up where available.
What logistics contacts can contribute
People who regularly visit industrial parks may know where a business is located and how to pass on a commercial invitation. That provides a possible entry point. It does not make a courier an industry expert.
A business should voluntarily respond and decide what information it can share. Photography, records, sampling and introductions require permission for the agreed task. Professional factory audits, testing and certification should be performed by the appropriate specialists. A few photographs cannot serve as a quality guarantee.
Start with one request in Shenzhen or Dongguan
The original article proposed connecting 100,000 logistics stations and one million Chinese factories. These are long-term goals, not the size of an operating network. The proposed starting point is Shenzhen, Dongguan and contacts I already know.
One useful engagement might identify a suitable business, confirm the responsible person and advance a site inquiry, quotation or sample. The record should capture the contact, permission, findings and next step. Adding company names alone does not complete a sourcing task.
Divide the work between people and AI
- AI helps interpret requirements, organize materials, shortlist leads, translate and compare.
- People handle agreed tasks that require local contact or presence.
- Records state the date, source, permission and unresolved questions so the buyer can continue evaluating.
Each observation has a limited scope and lifespan. Confirming an address or obtaining a sample does not confirm every production capability, certification or batch.
Connect supplier information with procurement follow-up
The original article connected supplier promotion with buyer assistance: one side explains products and capabilities; the other finds the business, responsible person and next action. The project aims to answer who can help, how to contact them and what happens next.
The project website is AnythingChina.net. Available locations, site tasks, records and fees need to be agreed for each engagement. Planned network scale should not be presented as current service capacity.
This follows my approach to embedded AI implementation: information should support action, with people responsible for both the technical and business work.
By Hequbing (He Fangsheng / Dosen). English edition prepared on 2 October 2026, based on my WeChat article (Chinese). Original illustrations and publication records are available there.
Contact: WeChat hecare888; email hefangsheng@gmail.com. Service scope and pricing.