At Intel Corporation, I owned data gathering, model training, evaluation, and production deployment for edge-deployed audio recognition and sound classification models. I experimented with CNN and Transformer architectures, with over 8 models in production.
At Amazon Development Center, I designed and built automated Python post-processing pipelines that reduced manual speech-data processing time by over 50%. I also managed acquisition and quality control for a large-scale multilingual speech corpus used to develop Amazon Alexa and AWS Polly voices.
I’ve worked across the end-to-end machine learning lifecycle, from data cleaning and validation to production deployment. In both roles, I served as Scrum Master for a five-person team; at Amazon, I also led and mentored two junior audio engineers.
