Sérgio Valentim

Platform development focusing on speech analysis features to help users improve English pronunciation, fluency, and speaking skills.

Software engineer delivering backend services and ML pipelines; improved real-time ML inference pipelines and reduced file handling times through optimized processing; designed and deployed containerized microservices on AWS with Docker and Kubernetes; established unified observability and automated testing in GitLab CI/CD; contributed to reducing infrastructure costs and increasing system reliability.

Backend engineer contributing to Elsa Speak platform; designed and implemented ML inference pipelines using Python, PyTorch, and TensorRT; reduced service latency with asynchronous processing; improved data annotation workflow; built scalable CI/CD pipelines; enabled rapid adoption of new ML techniques.