AI Intern at Innomatics Research Labs

Implemented fine-tuning workflows on transformer models using the LangChain framework, building custom intelligent agents with retrieval-augmented generation (RAG), memory, and tool-calling capabilities. Developed RAG pipelines integrating vector databases such as FAISS with OpenAI APIs for contextual question answering on large clinical datasets. Designed prompt templates and optimized workflows to enhance LLM relevance and efficiency for domain-specific applications. Deployed AI microservices using Docker and AWS Lambda, and implemented CI/CD pipelines through GitHub Actions to improve system reliability and monitoring.

Data Science Intern at Coapps Development Solutions India

Fine-tuned LSTM and GRU deep learning models for sales forecasting on Azure ML, achieving a 12% accuracy improvement. Built interpretable fraud detection pipelines using PyTorch focusing on explainable machine learning. Developed interactive Power BI dashboards for image analytics and monitoring key product metrics.

AI Intern at Innomatics Research Labs

Implemented fine-tuning workflows on transformer models using LangChain framework; built custom intelligent agents with RAG, memory, and tool-calling capabilities. Developed RAG pipelines integrating vector databases (e.g., FAISS) with OpenAI APIs for contextual question answering on large clinical datasets. Designed prompt templates and optimized workflows to improve LLM output relevance and efficiency for domain-specific applications. Deployed AI microservices with Docker and AWS Lambda and implemented CI/CD pipelines with GitHub Actions.

Data Science Intern at Coapps Development Solutions India

Fine-tuned LSTM and GRU models for sales forecasting on Azure ML, improving accuracy by 12%. Built interpretable fraud detection pipelines using PyTorch. Created interactive Power BI dashboards for image analytics and monitoring key product metrics.

AI Intern at Innomatics Research Labs

Implemented fine-tuning workflows on transformer models using LangChain framework built custom intelligent agents with RAG, memory, and tool-calling capabilities. Developed RAG pipelines integrating vector databases (e.g., FAISS) with OpenAI APIs for contextual question answering on large clinical datasets. Designed prompt templates and optimized workflows to improve LLM output relevance and efficiency for domain-specific applications. Deployed AI microservices with Docker and AWS Lambda and implemented CI/CD pipelines with GitHub Actions, improving reliability and monitoring.