MOHAN SRINIVAS

Generative AI Engineer with 2.5 years of experience building scalable AI-powered applications using Large Language Models (LLMs). Strong expertise in Python, SQL, REST APIs, and backend AI system development. Designed and deployed Retrieval Augmented Generation (RAG) pipelines for enterprise knowledge systems. Hands-on experience in Prompt Engineering, optimization, structured prompting, and prompt versioning. Worked extensively with OpenAI, Azure OpenAI, Hugging Face, and AWS Bedrock models. Implemented vector search using FAISS, Pinecone, ChromaDB, and Weaviate for semantic retrieval. Built AI-powered chatbots, document Q&A platforms, AI search engines, and summarization systems. Integrated AI systems with PDFs, CSVs, SQL databases, APIs, and AWS S3 storage. Deployed AI services using AWS Lambda, API Gateway, Step Functions with monitoring and logging. Focused on cost optimization, performance tuning, security, and scalable AI architecture.

• Exploring LLM fine-tuning and parameter-efficient adaptation techniques.
• Learning agent-based architectures and multi-agent AI systems.
• Gaining exposure to multi-modal models (text, image, audio).
• Studying Responsible AI, governance, and bias mitigation practices.

Client: Verizon Communication | Domain: Telecom

Client: ThinkBio.AI | Domain: Healthcare

· Designed end-to-end Gen AI solutions aligned with business requirements.

· Developed RAG-based document Q&A systems for large enterprise datasets.

· Built AI-powered chatbots and search tools improving operational efficiency.

· Integrated enterprise data sources including PDFs, CSVs, APIs, and SQL databases.

· Implemented logging, monitoring, performance tuning, and security best practices.