Profile

With a background in Computer Science and Technology, I am experienced with large language model applications and deep learning frameworks. During my internship at Notta, I focused on LLM application development and production-oriented microservices, with document generation as my primary responsibility, while also contributing to ASR tuning, image-generation model evaluation, and voice editing features. I am now seeking an AI application development or algorithm engineering role in Hong Kong, with a focus on turning model capabilities into reliable products.

Education

1. 2026 – 2027
   The Hong Kong Polytechnic University, Hong Kong.
   M.Sc. in Blockchain Technology.
2. 2022 – 2026
   Shenzhen Technology University. B.Sc. in Computer Science and Technology (Top 5%); 2025 Learning Star, First Prize (Top 3%).

Experience

1. Nov 2025 – Jul 2026
   Notta | AI Meeting Notes & Audio Transcription AI Algorithm Intern

   • Primary responsibility: Led the development of an LLM-powered document-generation microservice for Word, Excel, and PowerPoint outputs, using XML-level ReAct tool calls to plan, edit, and validate structured documents; the new version improved both cost and speed by 50% over the previous version and received positive feedback from multiple Japanese B2B customers.
   • Contributed to ASR algorithm development and tuning, including recognition-quality analysis, speech model optimization, and automated evaluation pipelines for repeatable speech algorithm testing.
   • Notta Brain infographic generation: designed benchmarks and test cases to evaluate leading image-generation APIs, including Qwen, Seedream, Nano Banana, and GPT Image 2, tracking generation time and cost through visualized comparisons; packaged the evaluation workflow as a Skill for automated testing of new models.
   • Speakon Edit voice correction: selected models from the candidate pool and designed prompts to evaluate and optimize the voice-editing experience.

Skills

Proficient in Python and PyTorch for deep learning development, with hands-on experience in Prompt Engineering, Agent / Tool Calling (ReAct), and the Agno framework for LLM applications. Experienced in packaging model capabilities as microservices with Docker and FastAPI, as well as structured document generation for Word / Excel / PowerPoint and ASR development and tuning. Proficient with Vibe Coding tools including Codex, Cursor, Claude Code, Kiro, and OpenCode.

Selected Publications

1. Wenbin Feng, Yu Lu, Xiaoqing Li, Kai Leung Yung, Wai Hung Ip. “Fog/Edge-Aware State Space Models for Multi-Task Chest X-ray Report Generation and Lesion Detection.” IEEE Journal of Biomedical and Health Informatics. Accepted / Early Access, 2026. DOI: 10.1109/JBHI.2026.3667969. (JBHI, JCR Q1) [IEEE Xplore]
2. Huilin Ge, Jie Zhu, Wenbin Feng, Jiali Ouyang, Ze Wang, Xiaoping Chen, Yu Lu. “AquaSlot-SAM: Coupling Slot-Based State Space Models with SAM for Robust Underwater Video Multi-Object Segmentation.” Pattern Recognition. Vol. 174, Article 112964, 2026. DOI: 10.1016/j.patcog.2025.112964. (Pattern Recognition, JCR Q1) [ScienceDirect]
3. Wenbin Feng, Yu Lu, Xiaoqing Li, Shijie Shi, Yingjian Qi. “MambaXray-CTL: Multi-Stage Contrastive Training for Medical Report Generation with a Mamba-Based Multi-Modal Large Model.” Proceedings of the IEEE International Conference on Systems, Man, and Cybernetics. 2025. DOI: 10.1109/SMC58881.2025.11343574. (IEEE SMC, Conference Paper) [IEEE Xplore]
4. Xiaoqing Li, Wenbin Feng, Yu Lu, Judice Koh, Ellie Choi, Jianli Chen, Jinhong He, Kee Yuan Ngiam. “Unsupervised Adaptive Path Optimization for Knowledge Graph Reasoning in Multimodal Medical Diagnosis.” Proc. 22nd International Conference on Intelligent Computing (ICIC 2026), Toronto, Canada. 2026. DOI: 10.1007/978-981-92-3441-7_39. (ICIC 2026, Conference Paper) [Springer]
5. Wenbin Feng, Yu Lu, Shijie Shi, Meng Li, Huilin Ge. “A Deep Learning Model for Surface Defect Detection in Thermoelectric Cooler Components.” Proc. International Conference on Intelligent Computing (ICIC), Lecture Notes in Computer Science: Advanced Intelligent Computing Technology and Applications. pp. 163–172, 2025. DOI: 10.1007/978-981-96-9921-6_14. (ICIC 2025, Conference Paper) [Springer]

Languages

Mandarin, Cantonese, English