Mastering Generative AI: From Prompts to Production with OpenAI, RAG, and AI Agents

BIO:

Hello, my name is Rohit Malla, and I am an AI Developer with over 1.9 years of experience in building intelligent, scalable software systems using cutting-edge technologies such as Python, FastAPI, LangChain, Hugging Face, and OpenAI APIs. My core expertise lies in designing and deploying LLM-powered applications, developing robust backend APIs, and integrating NLP models into real-world use cases to deliver business value. My journey in technology started with a passion for problem-solving and a deep interest in how machines understand human language. This curiosity led me to specialize in Natural Language Processing (NLP) and Large Language Models (LLMs). Over the past two years, I’ve been working at Trianz as a Software Engineer Trainee, where I had the opportunity to work on high-impact AI projects that combine backend engineering with AI development. One of the key highlights of my experience is developing an AI-powered code generation assistant using LangChain and GPT-based models, integrated with pgvector and PostgreSQL. This tool followed the Retrieval-Augmented Generation (RAG) pattern, leveraging custom embeddings and semantic search to provide developers with smart code suggestions. I implemented fallback mechanisms, prompt engineering strategies, and monitoring tools to ensure both accuracy and stability, reaching over 85% accuracy in generated code recommendations. On the backend, I specialize in building high-performance REST APIs using FastAPI and asynchronous programming patterns. I’ve deployed containerized microservices with Docker, managed secure authentication using OAuth 2.0 and JWT, and optimized query performance in PostgreSQL and MSSQL environments. My work has led to up to 40% improvements in API response times, and I’ve implemented robust logging and monitoring solutions using the ELK stack. In terms of AI application development, I’ve built tools such as: PDF AnswerBot – A document intelligence platform that extracts accurate answers from unstructured PDF documents using LangChain, OpenAI, and RAG with a Streamlit interface. SQL Genius – A natural language to SQL query converter that leverages Gemini Pro LLM and Python, enabling users to ask data-related questions and get optimized SQL queries in return. AI Code Assistant – A full-stack code suggestion tool that helps developers write code faster with real-time, context-aware suggestions based on their queries. I’ve also worked on enhancing search performance using Elasticsearch, vector embeddings, and Redis caching, all while maintaining high-quality standards through unit testing (Pytest, unittest) and CI/CD pipelines using GitHub Actions. I routinely maintain 90%+ test coverage in my projects and lead code reviews and API documentation efforts. Beyond coding, I’ve served as a Team Lead for AI projects, mentored junior developers in LLM integrations and Python best practices, and held key positions like Documentation Lead and core member of my college's tech club. I’m always researching new technologies and following developments in AI, especially around agent frameworks like LangGraph and orchestrators for multi-agent systems. I hold a B.Tech from Chandigarh University and several technical certifications, including: Advanced Java Programming (Internshala), DSA Specialization (Internshala), Google-C Certification (IIT Bombay), Generative AI with LangChain. In summary, I am a passionate AI Developer focused on building real-world, impactful solutions using modern AI technologies. I enjoy working in collaborative teams, solving meaningful problems, and continuously learning to stay ahead in this fast-paced field. Thank you for taking the time to know about me, and I look forward to opportunities where I can contribute to innovative AI projects and make a difference.