Nishvanth Thota

About Nishvanth Thota

AI models don’t just need fine-tuning—they need engineering that makes them work in the real world. As an LLMOps Engineer, I don’t just deploy models—I ensure they scale, optimize, and deliver impact. With expertise in AWS SageMaker, Amazon Bedrock, Amazon Q, RAG, RAPTOR, Evaluation using GISKARD and Hugging Face Transformers, I specialize in turning cutting-edge AI research into production-grade solutions.In just 1.8 years, I’ve helped optimize large-scale AI deployments, cutting inference costs by 35% and boosting retrieval efficiency by 50%. From fine-tuning multimodal GenAI models to building high-performance RAPTOR pipelines, my work drives AI-powered businesses forward. But for me, AI is more than just code—it\'s about solving real problems. I thrive on bridging the gap between research and real-world AI adoption, ensuring models are not just accurate, but also efficient and scalable. Looking for an LLMOps expert who can design, optimize, and scale AI systems for maximum business impact? Let’s connect. Gmail: n••••••••@gmail.com

Experience

1. Associate Software Engineer

   Tech Mahindra

   Dec 2023 — Present · Hyderabad, IN

   1) Designed and deployed scalable LLMOps pipelines, ensuring seamless AI model deployment, fine-tuning, and optimization for enterprise-scale applications2) Led efforts to reduce inference latency by 35%, slashing computational costs by 50%, while boosting retrieval efficiency for high-performance Key Achievements & Impact 1) Optimized LLM inference by accelerating response times by 70%, improving system throughput for large-scale AI deployments. 2)Implemented high-precision RAG pipelines, enhancing knowledge retrieval accuracy by 70%, improving chatbot response quality. 3) Developed RAPTOR-powered search algorithms, refining search precision and increasing retrieval relevance by 60%.4) Reduced model deployment time by 45%, leveraging AWS SageMaker and Amazon Bedrock for faster AI lifecycle management. 5) Enhanced multimodal GenAI models, improving cross-modal reasoning and vision-language model efficiency.6) Integrated FAISS for vector search, cutting data retrieval times by 3X optimizing AI-powered engines.

Education

• Sri Indu College Of Engineering & Technology

  Bachelor of Technology - BTech

  2019 — 2023