Sunil Thapa (Sunil)

- Over 4+ years of experience in AI Engineering with a strong focus on Natural Language Processing (NLP), Machine Learning, and deploying AI-driven solutions. 
- Expertise in developing and optimizing LLM-based systems, with proven ability to implement innovative approaches like retrieval-augmented generation (RAG) to enhance NLP tasks. 
- Strong proficiency in cloud infrastructure management (AWS, GCP, Azure) and ensuring scalable, fault-tolerant environments for AI models and systems. 
- Experienced in building and automating robust data pipelines, leveraging big data technologies such as Hadoop, Spark, and Hive to streamline data processing and analytics workflows. 
- Skilled in applying a wide array of machine learning techniques to solve real-world challenges, including building recommendation systems, face-attendance systems, and real-time object detection models. 
- Specialized in NLP solutions, including sentiment analysis, document summarization, and building multilingual chatbots, leading to improved user engagement and experience. 
- Proficient in deploying CI/CD pipelines with Terraform and GitHub Actions, significantly enhancing deployment efficiency and reducing system downtime. 
- Passionate about research and innovation, with published work on NepaliBERT and other cutting-edge NLP models, contributing to advancements in language processing for underrepresented languages. 
- Adept at collaborating with diverse teams to deliver AI-driven solutions that align with business goals and exceed client expectations. 
- Strong communicator with an emphasis on data-driven decision-making, using data visualization tools like Tableau and Power BI to present actionable insights. 
- A proactive mentor and team player, dedicated to continuous learning and enhancing the technical capabilities of teams.

As an AI Engineer at AgilePitch, I specialize in developing and deploying machine learning models, with a focus on Large Language Models (LLMs) and Retrieval-Augmented Generation (RAG) techniques. My work has led to a 25% improvement in model accuracy and a 40% increase in data processing speed. By integrating self-corrective RAG models, I improved factual accuracy in document retrieval by 28%. I have also optimized data pipelines, boosting data processing efficiency by 35%. Additionally, I have leveraged AWS services like S3, Lambda, and CloudWatch to enhance system reliability and operational efficiency by 13%. My work directly contributed to a 30% increase in model performance and overall customer satisfaction.