Big-Data Platform · Insolvency & Restructuring

Wupo Data (无破数据) · Big-Data Platform for Asset Restructuring

China's first structured insolvency data terminal—turning bankruptcy data scattered across judgments, announcements, and business-registry records into a searchable, subscribable, and callable industry infrastructure; with RAG (Retrieval-Augmented Generation) and AI Agent-based intelligent data retrieval, data access evolves from "writing query conditions" to "natural-language conversation".
0
Platform pre-registered users
0
Total funding raised (RMB)
0
WUPO Summit reach (person-times)
0
Institutional clients served
01
Background: From Casework Tool to Industry Data Foundation

The Wupo Digital Technology group operates two product lines: PoyiYun (破易云, a casework SaaS serving over a thousand institutions) and Wupo Data (无破数据). At the 2024 WUPO Summit, the Wupo Data terminal officially launched, marking the company's move from the tool layer into the data layer.

02
Product: Full-Dimension Insolvency Data Services

Covering bankruptcy-case data, enterprise data, document data, industry data, and institutional data; supporting advanced custom multi-dimensional search, deep cleansing and mining of data value, on-demand custom data fields (Excel / scheduled update pushes), and open API data interfaces for all stakeholder types. On top of this sit RAG intelligent Q&A and AI Agent data-retrieval services—users express needs in natural language and intelligently obtain the data they require.

03
Industry Influence

Since 2023, together with CNR.cn and XinhuaNet, the company has hosted the annual "WUPO Summit" and published the industry's first WUPO Directory, reaching over 50M person-times cumulatively; it has built data partnerships with AMCs (asset management companies), courts, and others to advance information flow across the distressed-assets industry.

01
Data Structuring Pipeline

Proprietary in-house data-structuring technology + AI algorithms: extracting case elements (parties, claim amounts, disposition status) from unstructured documents and announcements to build a continuously updatable structured database—a classic ETL + NLP engineering combination.

02
Multi-Dimensional Search and Data Mining

Supporting precise search with arbitrary condition combinations and cross-dimensional data correlation analysis, providing the data foundation for opportunity capture, trend analysis, enterprise-value identification, and asset-valuation trading scenarios.

03
API and Security & Compliance

Open API services for different stakeholders require a highly available API gateway, quota, and authentication system; meanwhile, full-lifecycle data-security management spans collection, storage, processing, and transmission.

04
RAG (Retrieval-Augmented Generation) and AI Agent-Based Intelligent Data Retrieval

A RAG pipeline built on structured data and document corpora: vector embeddings + multi-dimensional hybrid retrieval + result re-ranking, ensuring answers trace back to specific data entries. On top of this, an AI Agent data-retrieval assistant was developed—users state data needs in natural language, and the Agent automatically performs intent understanding, query-condition orchestration, cross-data-source aggregation, and structured output, turning "finding data" from hand-written queries into conversational intelligent retrieval.

Capital and Market Validation

The company has completed 4 funding rounds totaling RMB 34M, with the latest being the Pre-A+ round in November 2025 (tens of millions of RMB, with participation from Suzhou Asset Management); within its "AI tools + expert network + standardized services" product matrix, PoyiYun 4.0's open data platform now has over 450K pre-registered users. The group's mission—"to reshape the value of distressed enterprises and assets worldwide."

This project gave me deep hands-on practice with "data as a product": every link of collection, cleansing, and structuring is engineering work, and API-based output turns data into a continuously sellable service—methodologically identical to the AI engineering (RAG data pipelines) I do today.