Jingmin Zhou

Developing ML models for Foreign Materials (FMs) detection in food industry:

• Designed and developed robust ML models to meet customers' requirements for functionality, scalability, and performance
• Explored new features and optimized model performance: detection rate improved by 90% for a hard FM
• Led 2 major projects for model building and delivery by effectively managing time and prioritizing tasks according to importance
• Documented and communicated the results locally / to the cloud, other teams and the customers
• Mentored the new members on the model building track

Audio Transcript Mining using ML and NLP techniques

• Created Word Error Rate (WER) tool for evaluating the quality of transcripts
• Collaborated with colleagues to ensure seamless test and integration of new tools
• Programmed with Python data and ML libraries
• Did code review, debugging, deploying and documenting the users' guideline

Automated Video Categorization using Image and Audio Features

• Programmed with Python on high performance computing (HPC) platform, SOSCIP accelerated-GPUs.
• Implemented high-scaled experiments for classifying videos by visual and audio features..
• Gathered, arranged, and corrected research data to create representative graphs and charts highlighting results for presentations.
• As a speaker, presented the research work on IBM CASCON x EVOKE 2021