AI Prompt Engineer - AWS Bedrock
TextPrompt Response Writing SFT
As an AI Prompt Engineer at AWS Bedrock, I designed and evaluated prompts for large language model (LLM) outputs to improve accuracy in tasks such as summarization, classification, and reasoning. I performed red teaming using adversarial prompts and tested chatbot behaviors to ensure Responsible AI compliance. The work focused on enhancing conversational AI accuracy, safety, and alignment with guidelines.
• Designed and iteratively improved 50+ prompts for multiple task types.
• Carried out LLM output evaluations for intent, accuracy, and relevance.
• Performed red teaming with adversarial prompt scenarios to assess safety and output bias.
• Built agent workflows and benchmarked chatbot performance.
As an AI Prompt Engineer at AWS Bedrock, I designed and evaluated prompts for large language model (LLM) outputs to improve accuracy in tasks such as summarization, classification, and reasoning. I performed red teaming using adversarial prompts and tested chatbot behaviors to ensure Responsible AI compliance. The work focused on enhancing conversational AI accuracy, safety, and alignment with guidelines. • Designed and iteratively improved 50+ prompts for multiple task types. • Carried out LLM output evaluations for intent, accuracy, and relevance. • Performed red teaming with adversarial prompt scenarios to assess safety and output bias. • Built agent workflows and benchmarked chatbot performance.
2024 - Present