AI Link to heading

I work with Generative AI projects based on OpenAI models like GPT, integrating them with AWS based cloud applications. These projects add artificial intelligence features to web applications, allowing end-users to ask for help on using the application and to utilize the AI for several other purposes.

The systems that I have designed integrate with OpenAI in several ways, including calculating the cost of AI requests based on tokens, queuing and caching long running requests, applying default system prompt messages to customize the AI responses in a coherent way, training and using fine-tuned custom models for more detailed customizations, and applying tool integration to implement RAG (Retrieval Augmented Generation) by searching relevant content from Amazon OpenSearch.