Extensive hands-on experience in the development, deployment, and evaluation of large language models (LLMs) for practical, real-world applications.

Configured, deployed and served several Large Language models that perform machine translation, and text-to-speech (eg. NLLB, MMS, MATLAB400, MBART etc). This involves loading LLM from hugging face( or Openai), creating a Streamlit app, building a docker container for the model, pushing it to DockerHub and, finally running it on an AWS instance with GPU-enabled inferences.

Developed a conversational chatbot for Facebook Messenger by integrating AWS Lambda to handle API requests and responses, utilizing Neural Chat 7B to ensure seamless communication between ManyChat and Prediction Guard. Utilized Node.js for backend development, implementing custom logic for conversation flow and user interaction.