Build a Multilingual Logistics Intake and Cross-Border Transit Router

Parse unstructured shipping inquiries across English, Malay, Mandarin, and Tamil into normalized consignments. Evaluate port customs urgency and dispatch booking payloads to terminal operators.

Regional freight forwarder logistics coordinator

Ingest multilingual freight requests received through webhooks and messaging channels.

Translate text and extract normalized cargo dimensions, commodity codes, and arrival targets.

Score customs clearance urgency against Jurong and Tuas terminal berthing windows.

Dispatch structured logistics orders to terminal dispatch queues for operator confirmation.

Multi-pane routing console showing incoming raw messages, detected language confidence, normalized booking entity fields, and terminal dispatch status.

Create synthetic normalized inquiry datasets using real Data.gov.sg weather and traffic status schemas from source 0 for local transit context. Starter examples below are synthetic. They are not downloaded records.

Singapore open data API Queries real-time API formats referenced in data.gov.sg developer docs for traffic and weather context during transit routing.

pydantic-ai, fastapi, httpx. Pydantic AI ensures strict type validation on multilingual entity extraction before payload dispatch. Fast multilingual instruction-following model via hosted API for translation and structured entity extraction.

Submit runnable source, pinned dependencies, setup commands, a working interface, example outputs, and a recorded end-to-end demo. Include tests and an action log.

Correctly identifies English, Malay, Simplified Chinese, and Tamil message payloads.

Extracts pallet count, cargo weight, and destination terminal without schema errors.

Flags incomplete dangerous goods declarations before queue dispatch.

Dispatches normalized booking orders with deterministic urgency scores.

```json [ { "input": { "source_language": "ms", "raw_message": "Sila aturkan 14 palet barang elektrik dari Pasir Gudang ke Tuas Port sebelum 4 petang esok.", "sender_uen": "201823901M" }, "expected_output": { "detected_language": "ms", "normalized_order": { "cargo_type": "electrical_goods", "pallet_count": 14, "origin": "Pasir Gudang", "destination": "Tuas Port", "urgency_level": "standard", "routed_terminal": "tuas_gate_4" }, "status": "ready_for_dispatch" } }, { "input": { "source_language": "zh", "raw_message": "冷冻海鲜，明天送，数量没定", "sender_uen": "202011409K" }, "expected_output": { "detected_language": "zh", "normalized_order": { "cargo_type": "frozen_seafood", "pallet_count": null, "origin": null, "destination": null, "urgency_level": "blocked", "routed_terminal": null }, "status": "missing_required_dimensions" } } ] ```

Simulates all external carrier bookings and terminal gate dispatches without executing real-world commitments. Use mock external writes. Require human review before any real deployment.