MSXF (马上消费金融股份有限公司) | Chongqing/Beijing, China#

A Leading Consumer Finance Technology Institution in China

Director of AI Product / AI Product Expert | May 2016 – Present

• Head of AI Psychology Research Department | Feb 2025 – Present

  • Pioneered the application of AI & Cognitive Psychology Frameworks across 5 core business pillars: Debt Collection, Customer Service, Telemarketing, Recruitment, and Loyalty Programs.

  • Designed an AI-driven interview system that successfully evaluates candidate resilience, achieving a 9% absolute reduction in short-term employee turnover.

  • Architected high-EQ empathetic conversational frameworks based on an “8-category, 15-dimension psychological labeling system” ; enhanced agents’ empathetic speech adoption by 12.1%, driving a 0.74% absolute (16.45% relative) surge in telemarketing conversion rates.

  • Launched user-retention and mental wellness products, scaling user penetration to 2% with Daily Active Users (DAU) exceeding 10,000+.

• Head of Innovation Research Department | Apr 2021 – Feb 2025

  • Directed the development of the Multi-Modal Biometric Anti-Spoofing and Security Platform , defending against complex presentation attacks (deepfakes, 2D/3D printing, ROM injection, and link hijacking) with a >99% pass rate.

  • Scaled the platform to handle over 100 million annual API calls, safeguarding over ¥1 Trillion in financial assets.

  • Monetized proprietary security technologies through B2B commercialization via parent company, securing ¥3.36 Million in contracts from major tier-1 national banks (e.g., CITIC Bank Head Office, PSBC Head Office, China Merchants Bank Head Office).

  • Managed the Virtual Digital Human Framework, launching a self-service interactive 2D/3D digital avatar and content generation platform utilized by C-suite executives, renowned university professors.

• Head of AI Product Team – Intelligent Engineering Department | May 2020 – Apr 2021

  • Led the construction of foundational benchmarks for ASR, TTS, and Voiceprint datasets , drastically elevating baseline model deployment quality.

  • Secured over 10 authoritative capabilities certifications from national bodies including BCTC, AIIA, and the First Research Institute of the Ministry of Public Security.

• Head of Speech Recognition Department | Feb 2018 – May 2020

  • Built enterprise-grade AI Voice Conversational Robots (Telemarketing, Fraud Detection, and Quality Assurance) , supporting high-performance distributed architectures and business continuity.

  • Successfully managed operations supporting 1 Million+ daily outbound calls , 6,000+ daily hours of automated audio QA , and a peak concurrency of 10,000+ QPS.

• Core Algorithm Developer of Data Mining Department | May 2016 – Jan 2018

  • Engineered user communication network social graphs ($f(user)=topN (nodes)$) ; extracted N-gram and TF-IDF text features to train high-precision XGBoost models for loss recovery and cross-selling.

  • Achieved a 20% response rate in user acquisition (vs. a 6% baseline) , and directly recovered tens of millions of RMB in losses by lifting collection cure rates by 0.2%.

I. National & Association Standard Formulations (Selected)#

• Drafting Author, National Standard: Artificial intelligence—Computing center—Computing capacity assessment (GB/T 46346-2025 ).
• Drafting Author, National Standard: Information technology—Use of biometrics in video surveillance systems—Part 1: System design and specification GB/T 44261.1-2024).
• Drafting Author, Association Standard: Technical requirements of large model enhanced digital human system (T/BIA 30-2025).
• Drafting Author, Association Standard: Technical requirements and test methods for digital human Part 1: Platform basic competency (T/BIA17-2024).

II. Books & Academic Publications#

• Monograph Co-Author: Financial Large Language Models (Science and Technology China Press, ISBN: 9787523605578). Solely authored the “Security LLM” section in China’s first professional monograph dedicated to Financial LLMs.
• D. Zeng, L. Gao, H. Fang, G. Xiang, Y. Feng* and Q. Lu, “Bandpass Filter Based Dual-stream Network for Face Anti-spoofing,” 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), Vancouver, BC, Canada, 2023, pp. 6403-6410, doi: 10.1109/CVPRW59228.2023.00681.
• N. Jiang, D. Zeng, L. Gao, S. Chen, Z. Kong, Y. Liu, J. Li, Y. Feng, T. Yuan, W. Deng, “MS-UFAD: A Large-Scale Dataset for Real-world Unified Face Attack Detection with Text Descriptions,” ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Hyderabad, India, 2025, pp. 1-5, doi: 10.1109/ICASSP49660.2025.10887853.
• N. Jiang, Y. Liu, D. Zeng, Y. Feng, W. Deng and Y. Li, “Device-aware Optical Adversarial Attack for a Portable Projector-camera System,” ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Hyderabad, India, 2025, pp. 1-5, doi: 10.1109/ICASSP49660.2025.10889447.

III. Patents & Innovations (15 First-Inventor Filings out of 25+ Total)#

• Second Inventor, Data Extraction Method, Equipment, and Related Apparatus (Patent ID: CN111311041A).
• First Inventor, Liveness Detection Method, Device, Electronic Equipment, and Storage Medium (Patent ID: CN117953592A).
• First Inventor, Data Processing Method, Device, Electronic Equipment, and Storage Medium (Patent ID: CN11791171A).

IV. Global Technical Competitions (As Team Leader & Advisor)#

• Grand Champion (Dual Tracks), 2023 VoxSRC (International Voiceprint Recognition Competition).
• Grand Champion, FaceForensics++ Benchmark (Deepfake Global Academic Leaderboard).
• 2nd Place, 2023 ADD (Audio Deepfake Detection Challenge).
• 3rd Place, 4th Face Anti-Spoofing Challenge hosted by CVPR Workshop 2023.