Architected agentic AI platform using AWS Bedrock and Terraform reducing environment setup time by 70%

Engineered 5+ production RAG chatbots with LLMs and vector embeddings handling 10K+ daily queries with 95% user satisfaction

Led end-to-end ML deployment using Docker and CI/CD implementing monitoring, A/B testing, and auto-scaling for distributed systems

Enhanced auto-annotation workflows integrating LLMs and active learning reducing manual labeling by 60%

Implemented Git version control and CI/CD pipelines improving code quality and test coverage to 85%

Accelerated dataset creation by 4 weeks per iteration across 4 engineering teams

Developed sales forecasting models using ARIMA, Prophet, and LSTM increasing prediction accuracy by 30%

Automated data-entry using OCR and Selenium eliminating 100% manual entry and improving accuracy to 98%

Saved $50K+ in operational costs through improved forecasting accuracy