PandaAI
Category: AI Agents
Chinese AI trading infrastructure startup building an AI-agent-driven quantitative trading platform that turns trading ideas into verifiable strategies via natural-language workflows. PandaAI was founded in 2024. The company is led by 李昱琦 (Li Yuqi). Based in Chongqing, China. Team size: 51-100. Total funding raised: $4.5M. Latest round: Angel. Key investors include L2F 光源创业者基金 (L2F Lighthouse Founders' Fund).
- Founded
- 2024
- Headquarters
- Chongqing, China
- Team size
- 51-100
- Total funding
- $4.5M
Value proposition
Democratizes institutional-grade quantitative trading research by using AI agents to orchestrate the full pipeline — data acquisition, factor construction, strategy generation, backtesting, risk assessment, and live trading execution — so individual traders and small teams can compete with large quant funds.
Products and solutions
Qube (entry-level natural-language strategy generation & backtesting), EVO (AI-native quant research workbench for professionals, "quantitative Claude Code"), OS (A2A multi-agent collaboration platform, upcoming), QuantSkills (open-source quant capability library), TQX (overseas version), PandaData (data service), factor competition platform.
Unique value
The moat is orchestration (Harness layer) — combining models, data, QuantSkills, agents, and constraint rules — not the models or data themselves, which can be bought. Enables anyone to build and manage a team of trading agents.
Target customer
Individual retail traders (95后-05后 AI-native users), professional quant researchers, securities/futures firms, funds, and financial institutions.
Industries served
Quantitative finance / algorithmic trading, asset management, financial technology, securities and futures.
Technology advantage
Self-developed CQ2 neuro-symbolic LLM agent for quant research; A2A (Agent-to-Agent) multi-agent architecture; ADE (Agent Development Environment); integrates DeepSeek and Doubao external models; published papers (CQ2, A2A Self-Evolution, AlphaSchema); open-sourced QuantSkills and agent cluster (June 2026); full futures trading counter connectivity with live trading client launched June 2026.
How they differentiate
Focuses on AI-agent orchestration (Harness layer) rather than model/data advantage; emphasizes multi-agent collaboration and self-evolution; open-source QuantSkills ecosystem; targets individual traders to democratize institutional quant research capability; positions as "quantitative Claude Code" serving individuals directly rather than selling APIs to institutions.
Main competitors
同花顺 SuperMind (quant platform), 东方财富妙想 (East Money AI research), BigQuant (AI quant investment platform), Northstar (open-source AI quant trading platform).
Key partnerships
L2F 光源创业者基金 (L2F Lighthouse Founders' Fund, lead investor), partnered with 国泰海通证券重庆分公司 (GuTai Haitong Securities Chongqing) and 宏源期货 (Hongyuan Futures) for factor competition summit, 格林大华期货 (Green Da Hua Futures) cooperation for account opening, collaborations with securities firms, futures companies, funds, and financial institutions.
Notable customers
Securities firms, futures companies, funds, and professional financial institutions (specific names not publicly disclosed), 国泰海通证券重庆分公司 and 宏源期货 co-hosted factor summit.
Major milestones
Founded 2024, first external funding H2 2025 (3 rounds in 2 months), open-sourced QuantSkills & Agent cluster June 2026, live futures trading client launched June 2026, 3rd factor competition & AI Trading Summit held Shanghai Aug 2026, published CQ2, A2A Self-Evolution, AlphaSchema papers.
Growth metrics
100K+ users domestically and internationally; 70+ employees; 20K+ registered users on overseas TQX version; 3 factor competitions with 30K total participants (3rd competition 12K+ registrants, 8K+ factors submitted).
Market positioning
Early-stage Chinese AI trading infrastructure startup; first mover in AI-native agent-driven quant research for individual traders; building a "new Trading Infrastructure for the AI era" with a growing quant community (100K+ users, 30K factor competition participants).
Geographic focus
China (domestic) plus overseas expansion via TQX product targeting Southeast Asia and Middle East markets lacking mature quant infrastructure.
Patents and IP
Published research papers: "PandaAI: A Practical Agent CQ2 for Neuro-symbolic Data Analysis And Integrated Decision-Making in Quantitative Finance" (arXiv 2606.06823); A2A Self-Evolution; AlphaSchema.
About 李昱琦 (Li Yuqi)
Ex-quant hedge fund partner (managed >¥1B assets); Columbia University BS+MS in Financial Engineering; founder of quant finance education IP "量化李不白" (200K+ followers); started quant private equity entrepreneurship at age 20.
Latest news about PandaAI
More AI Agents companies
Official website: https://www.pandaaiquant.com/
