aguang

I am an AI engineer with 10+ years of hands-on experience designing and delivering end-to-end algorithm systems across computer vision, NLP, speech (ASR/TTS), 3D reconstruction and large-scale LLM/agent applications. I have led projects from problem definition and data acquisition through model design, training pipelines, quantization and on-prem deployment. Selected project highlights include medical imaging and 3D reconstruction pipelines, multilingual document RAG systems, multi-role emotional TTS, large-model fine-tuning with quantized local serving, multi-agent orchestration and production automation that connects ASR → RAG → business workflows. 

Competition & awards

Kaggle ML competitions — Silver & Bronze placements. 

EmotW / ICMI Attention Detection Competition — 2nd place (2018, 2019, 2020). 

FME Micro-expression Recognition Challenge — 3rd place (2021). 

Selected publications

Multi-feature and Multi-instance Learning with Anti-overfitting Strategy for Engagement Intensity Prediction. 

Facial Action Unit-based Deep Learning Framework for Spotting Macro- and Micro-expressions in Long Video Sequences.  

Technical stack (representative)
 PyTorch, TensorFlow, Keras, MONAI, Hugging Face Transformers, PEFT, TRL; OpenCV, OpenPose, OpenFace; COLMAP, NeRF / 4D Gaussian Splatting, point-cloud tooling; Docker, Ollama, vLLM and related LLM serving tools.

I am seeking senior technical roles such as Senior/Lead AI Algorithm Engineer, Research Scientist, or ML Architect.