Srinidh Vu

About Srinidh Vu

I am a senior GenAI and Machine Learning Engineer with over 10 years of experience architecting, building, and deploying production-grade AI solutions. My expertise bridges the gap between cutting-edge research and tangible business value, delivering systems that are scalable, reliable, and intelligent. I specialize in the end-to-end AI lifecycle—from large-scale data engineering to MLOps and the fine-tuning of Large Language Models (LLMs). Core Expertise GenAI & Deep Learning: Large Language Models (LLMs), RAG, Fine-Tuning (PEFT, LoRA/QLoRA), prompt Engineering, Transformers, Hugging Face, PyTorch, TensorFlow. MLOps & Deployment: LLMOps/MLOps Frameworks, Kubernetes, Docker, CI/CD (Azure DevOps, Jenkins), Terraform, Kubeflow, MLflow. Data Platforms & Big Data: Databricks, Apache Spark, Kafka, Feature Stores, Real-time Streaming analytics. Cloud Platforms: Azure Cloud, AWS Cloud (including SageMaker, S3, Glue). What Sets Me Apart? End-to-End GenAI Specialist – Proven experience leading the full lifecycle of Generative AI projects. I\'ve architected multi-agent RAG systems that cut research time by 60% and fine-tuned proprietary LLMs that outperform GPT-4 on domain-specific tasks. Production & MLOps Expert – I don\'t just build models; I make them work at scale. I\'ve deployed high-throughput services on Kubernetes handling over 1 million daily inferences and built complete MLOps pipelines to ensure 99.9% uptime and monitor for model drift.🧠 Advanced Modeling & Problem Solving – I apply state-of-the-art techniques to solve the toughest challenges. My work includes developing Graph Neural Networks (GNNs) that improved complex fraud detection by 7% and multimodal models for analyzing text and image data.🤝 Technical Leader & Mentor – Experienced in leading AI initiatives within Agile squads, mentoring junior engineers, and collaborating with business stakeholders to translate complex requirements into technical solutions. Why Connect With Me? Need a senior engineer to lead the architecture and deployment of your GenAI or LLM applications? Looking to productionize your machine learning models with robust, scalable MLOps pipelines? Interested in leveraging state-of-the-art AI to solve complex business problems in finance or healthcare? Let’s connect to discuss how I can help your team build the next generation of intelligent systems. s••••••••@gmail.com

Experience

1. Generative Ai Engineer

   Discover

   Oct 2023 — Present · Phoenix, AZ, US

   Architected a multi-agent RAG system using LangChain & Neo4j, cutting financial analyst research time by 60%. Adapted large foundation models (LLMs, VLMs) for specialized downstream tasks using advanced fine-tuning (PEFT - LoRA/QLoRA) & prompt engineering, improving domain-specific performance. Optimized and adapted a Llama 3 model with QLoRA quantization, slashing model size by 75% while boosting inference speed 2x. Deployed a high-throughput personalization service on Kubernetes, handling over 1 million daily LLM inferences. Implemented a full LLMOps pipeline to monitor for data & prompt drift in production, reducing chatbot escalations by 20%. Architected multimodal computer vision pipelines on AWS (SageMaker, S3, Glue) integrating visual and associated textual data for enhanced analysis and fraud detection.🧠 Engineered the data pipeline to continue pre-training a Llama 3 model on a 50B token corpus, creating a proprietary model that beat GPT-4 by 12%. Utilized a LayoutLM model. to parse complex PDF financial documents (combining layout/vision and text). automating extraction with 99% accuracy Developed a custom evaluation framework to benchmark LLMs on proprietary data, leading to a 15% improvement in model selection accuracy.🤝 Integrated a RAG system for internal HR & compliance documents, reducing employee clarification emails by 50%.

Education

• Jawaharlal Nehru Technological University, Kakinada

  Bachelor of Technology - BTech, Computer Science