Tradeics uses AI (alongside blockchain and cloud infrastructure) to make B2B trade faster and more transparent — assisting teams across procure-to-pay, sales, and operational decisions.
Where AI sits in the architecture
AI features operate on top of platform data. Clean masters and transactional history make recommendations useful; dirty ERP mirrors limit value.
ERP / CRM ──sync──► Tradeics B2B Platform ──► Tradeics AI assists
│
└──► Tradeics Finance settles
Your app ──HTTPS──► https://llm.tradeics.example/v1 ──► modelsTwo ways to use Tradeics AI
| Path | When to use | Entry point |
|---|---|---|
| Product AI | Recommendations and assists inside Tradeics workflows | Capabilities |
| LLM API | Call models from your backend with OpenAI-compatible SDKs | Call the LLM · LLM API Reference |
Quick LLM example
curl https://llm.tradeics.example/v1/chat/completions \
-H "Authorization: Bearer $TRADEICS_LLM_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "<model_id>",
"messages": [
{"role": "user", "content": "Summarize this RFQ for a buyer."}
]
}'Developer implications
- Prefer structured fields (categories, units, Incoterms, currencies) over free-text only payloads.
- Keep stable identifiers — AI outputs should reference Tradeics object IDs your systems already store.
- Treat AI suggestions as assistive: require human or policy confirmation before irreversible commits (PO award, payout release) unless a product API explicitly defines auto-apply modes.
What this launch documents
- LLM inference routes on
https://llm.tradeics.example(chat, models, embeddings, and more) - Product AI capability areas to design around while workflow APIs expand
- Platform REST V2 for masters and upcoming product routes