Navin Mourya

Experienced in building LLM-powered systems, RAG pipelines, and AI agents using LangChain, OpenAI, and AWS Bedrock. Skilled in vector database optimization, prompt engineering, and cloud deployment (Lambda, EC2, S3). Passionate about scalable AI infrastructure and transforming R&D into production-grade tools.

AI Engineer And Prompt Engineer at TechHighway System Pvt ltd (2025-06 – Present)

Building LLM-powered systems, RAG pipelines, and AI agents with focus on scalable AI infrastructure

• Conducted extensive R&D on Retrieval-Augmented Generation (RAG) systems, focusing on database integration and leveraging Langchain Retriever and embedding models.
• Employed advanced retrieval techniques such as LOTR Retriever, BM25, and similarity-based methods to enhance RAG performance.
• Tested and evaluated various open-source LLMs locally using OLLAMA, including Llama 3.1, Phi 3, Mistral, and Gemma 2.
• Utilized AWS for service development, implementing Lambda functions and monitoring with CloudWatch.

AI Engineer at TechHighway System Pvt ltd (2025-06 – Present)

Developed RAG-based systems for deduplication and metadata enrichment

• Developed a RAG-based system for deduplication and metadata enrichment across a 13M-record dataset, achieving 95% matching accuracy and reducing editorial effort by 80%.
• Conducted R&D on LangChain retrieval systems with Qdrant, Milvus, and PGVector, improving retrieval efficiency by 30%.
• Built CI/CD pipelines using AWS CodePipeline and AWS Bedrock for automated deployment and inference workflows.

AI Engineer at TechHighway System Pvt ltd (2025-06 – Present)

Developed RAG-based hotel information chatbot supporting semantic search

• Developed a RAG-based hotel information chatbot supporting semantic search across hospitality datasets.
• Built data ingestion pipelines processing 8K+ hotel records daily for real-time retrieval.

AI Engineer at Webomates (2024-03 – 2025-05)

Conducted intensive research in LLM field and developed AI applications

• Conducted extensive R&D on Retrieval-Augmented Generation (RAG) systems, focusing on database integration and leveraging Langchain Retriever and embedding models.
• Employed advanced retrieval techniques such as LOTR Retriever, BM25, and similarity-based methods to enhance RAG performance.
• Tested and evaluated various open-source LLMs locally using OLLAMA, including Llama 3.1, Phi 3, Mistral, and Gemma 2.
• Utilized AWS for service development, implementing Lambda functions and monitoring with CloudWatch.
• Done a intensive research in the field of LLM(Large language model) model.
• Developed applications using Streamlit, Langchain, and Various opensources LLms, and deployed them on Google Cloud.
• Guided an AI engineering team effectively.
• Conducted rigorous testing and validation of an AI-powered chatbot.
• Oversaw vendors for making and launching apps.

AI Engineer at Elevatporium Technology (2023-07 – 2024-02)

Conducted intensive research in LLM field and developed AI applications

• Done a intensive research in the field of LLM(Large language model) model.
• Developed applications using Streamlit, Langchain, and Various opensources LLms, and deployed them on Google Cloud.
• Guided an AI engineering team effectively.
• Conducted rigorous testing and validation of an AI-powered chatbot.
• Oversaw vendors for making and launching apps.

B.Tech in Computer Science & Business System – Bharati Vidyapeeth Deemed to University department of engineering and technology (2023)

Diploma in Computer Technology – Bharati Vidyapeeth Deemed to University department of engineering and technology (2019)