Aashmit Shrestha

I am a highly motivated Software Engineer with a strong foundation in AI, particularly in Large Language Models (LLMs) and full-stack development. My journey has been marked by a proven ability to architect and implement production-grade Retrieval-Augmented Generation (RAG) systems, optimizing LLMs through fine-tuning and Reinforcement Learning from Human Feedback (RLHF). I take pride in building robust data ingestion pipelines and have a keen focus on performance, scalability, and system resilience.

Currently, I work as an Associate Software Engineer at Rumsan Associates, where I have successfully optimized LLMs to enhance contextual understanding and response relevance. My experience includes architecting multi-agent frameworks and developing secure FastAPI backends. Additionally, I have contributed to significant projects such as the UNICEF Giga Schools initiative, processing millions of data rows while ensuring system resilience and efficiency. I am also a published researcher, having presented my work on stock market prediction at an international conference.