VARSHA MP

I am a Computer Science and Engineering student with hands-on experience building AI, AR, and cloud-native systems. I blend research internships with practical projects to deliver applied solutions that bridge models and real-world interactions.

At IIT Kanpur I designed AR experiences using ARCore, Geospatial APIs, and GPS-based placement to create interactive campus tours with contextual responsiveness. At Anna University I developed an AI pipeline for Amazon review scraping and achieved a 76% improvement in feature-extraction accuracy using NLTK and SpaCy.

My projects include an enterprise AI agent on Amazon Bedrock that uses RAG, Titan embeddings, fine-tuned models, OpenSearch Serverless, and AWS Lambda orchestration to auto-generate reports and integrate with Slack. I also built EduAdapt, a personalized learning platform with learning-style classification and retrieval-based content generation, and a multi-agent code-generation system using LangGraph and Groq.

I work primarily with Python, PyTorch, HuggingFace, LangChain, AWS (including Bedrock and Lambda), and related data tools; I focus on reliable, deterministic AI pipelines, retrieval augmentation, and practical integrations that deliver measurable improvements.