Kando AI raises tens of millions of yuan in seed funding to build 'Cursor for decision-making'

Kando AI was founded in June 2026 by Wu Bingzhe (PhD in computer science from Peking University, former Tencent AI Lab researcher) and Mao Shuhan (Tsinghua University graduate, serial entrepreneur). The company has raised a seed round led by Xinglian Capital, with Lihe Jinkong and an industrial investor participating. Kando AI's product is a workbench for high-value decision-making, focusing on areas such as finance and scientific research. Unlike traditional AI tools that optimize output quality, Kando AI aims to improve decision adoption rate and post-hoc reliability. The system features a memory mechanism that goes beyond RAG, remembering not just what users saw but how they understood problems, their preferences, and past judgment corrections. It creates a recursive loop of 'judgment-action-feedback-update', where each decision outcome changes how the system retrieves information and generates future suggestions. The team positions this as 'generalized post-training', optimizing context, interaction, and memory rather than just model parameters. Kando AI has developed its own infrastructure layer for transparency and traceability. The product is currently in beta testing, targeting professional users like investors and researchers.