Date: Apr 2025 – Present
Roles: Machine Learning Engineer, MLOps Engineer
This project was completed while I was a consultant at Melio AI, and some of the project details have been obfuscated.
Overview
Developed an automated quality assurance system to evaluate call center agent compliance by transcribing audio and analysing conversations using language models. The system was designed to ensure auditability and transparency in regulated financial environments.
My Responsibilities
- Designed and implemented modular training and inference pipelines.
- Integrated model artefact lineage tracking for compliance and auditing purposes.
- Structured pipelines to support multi-model workflows, enabling reliable deployment and version control.
- Provided architectural guidance to enable the broader team to train and deploy models efficiently.
Outcomes/Impact
- Enabled scalable and transparent deployment of compliance-focused ML models.
- Improved the team’s ability to move models from development to production through reusable pipeline infrastructure.
Tools Used
- Python
- Docker
- AWS
- AWS Lambda
- AWS SageMaker
- HuggingFace Transformers
- OpenAI Whisper