Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

AI/LLM Engineer with 5+ years of experience architecting and deploying production-grade Generative AI, Large Language Model (LLM), and Machine Learning solutions for enterprise-scale applications. Experienced in building Retrieval-Augmented Generation (RAG) systems, fine-tuning and serving foundation models, designing AI agent workflows, and developing scalable inference platforms that process millions of requests with high availability and low latency.…AI/LLM Engineer with 5+ years of experience architecting and deploying production-grade Generative AI, Large Language Model (LLM), and Machine Learning solutions for enterprise-scale applications. Experienced in building Retrieval-Augmented Generation (RAG) systems, fine-tuning and serving foundation models, designing AI agent workflows, and developing scalable inference platforms that process millions of requests with high availability and low latency.

Saiteja Ramini





AI/LLM Engineer with 5+ years of experience architecting and deploying production-grade Generative AI, Large Language Model (LLM), and Machine Learning solutions for enterprise-scale applications. Experienced in building Retrieval-Augmented Generation (RAG) systems, fine-tuning and serving foundation models, designing AI agent workflows, and developing scalable inference platforms that process millions of requests with high availability and low latency.…AI/LLM Engineer with 5+ years of experience architecting and deploying production-grade Generative AI, Large Language Model (LLM), and Machine Learning solutions for enterprise-scale applications. Experienced in building Retrieval-Augmented Generation (RAG) systems, fine-tuning and serving foundation models, designing AI agent workflows, and developing scalable inference platforms that process millions of requests with high availability and low latency.

Available to hire

AI/LLM Engineer with 5+ years of experience architecting and deploying production-grade Generative AI, Large Language Model (LLM), and Machine Learning solutions for enterprise-scale applications. Experienced in building Retrieval-Augmented Generation (RAG) systems, fine-tuning and serving foundation models, designing AI agent workflows, and developing scalable inference platforms that process millions of requests with high availability and low latency.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Intermediate

Intermediate

Language

Work Experience

AI/LLM Engineer at Netflix

May 1, 2023 - Present

Architected enterprise-scale RAG platforms using LangChain, LangGraph, GPT-4o, Claude, and FAISS for semantic content discovery; improved recommendation relevance by 32% and reduced hallucinations by 28% via hybrid retrieval and metadata-aware filtering. Engineered highly available LLM inference services using FastAPI, Docker, Kubernetes, Redis, AWS Bedrock, and SageMaker, handling 10M+ daily requests with 99.95% availability and 38% lower latency through asynchronous execution and optimized GPU utilization. Built QLoRA fine-tuning pipelines for Llama 3 to improve domain-specific quality by 27% while reducing GPU memory and inference costs by 31%. Created distributed embedding/vector indexing pipelines with Kafka, Spark, Pinecone, PostgreSQL, and Redis, reducing indexing time by 42%. Established end-to-end LLMOps workflows with LangSmith, RAGAS, DeepEval, CI/CD automation, and Terraform to reduce regressions by 36% and improve release reliability. Developed multi-agent AI workflows wit

Software Engineer (AI/ML) at Cognizant

May 1, 2019 - July 1, 2021

Developed AI-powered fraud detection solutions using BERT, XGBoost, scikit-learn, PyTorch, and Python; improved detection accuracy by 41%, reduced false positives by 26%, and contributed to $2M+ annual savings. Engineered conversational AI using Rasa, Hugging Face Transformers, FastAPI, REST APIs, and MongoDB, supporting 50K+ daily interactions; improved first-contact resolution by 34% and customer satisfaction by 25%. Designed document intelligence pipelines with OCR, spaCy, BERT, and Airflow; reduced manual processing by 61% and improved extraction accuracy to 95%. Architected scalable ML pipelines on AWS SageMaker, Docker, Kubernetes, MLflow, and Flask for real-time credit risk models; reduced deployment cycles by 46% and improved prediction consistency by 29%. Optimized distributed ETL and feature engineering workflows using Spark, Kafka, SQL, Pandas, and Redis, improving processing speed by 39% and feature availability for downstream models. Automated deployment and infrastructure