LLM-Powered Marketing Applications
AI-driven applications at PMG built on AWS Bedrock and LangChain that streamlined data processing and personalized user experiences.
- Key results
- Improved marketing campaign targeting and customer engagement by 30%
- Resolved a cataloging bottleneck with SQS message queues, +40% throughput
- Optimized database schemas for 20% faster data retrieval
- Tech used
LLM-Powered Marketing Applications
At PMG I engineered AI-driven applications that leverage large language models through AWS Bedrock and LangChain. They streamline data processing and enable personalized user experiences, which improved marketing campaign targeting and customer engagement by 30%.
Beyond the LLM work, I found and fixed a critical bottleneck in the cataloging module by introducing message queues with AWS SQS, which improved system throughput by 40%. I also optimized database schemas and queries to get a 20% increase in data retrieval speed for client-facing applications.
It was the role where I learned to ship LLM features inside a real production codebase, with real users and real performance constraints.