Venkatesh Ramavath

About Venkatesh Ramavath

AI/ML Engineer with 5+ years of experience designing, training, and deploying large-scale machine learning models, LLMs, and real-time AI systems across global tech organizations. Skilled in PyTorch, TensorFlow, Hugging Face, and MLOps frameworks (MLflow, Kubernetes, Docker) with proven success in scaling NLP, computer vision, and multimodal AI solutions. Expertise in big data engineering with Apache Spark, FAISS, and Pinecone, enabling petabyte-scale data processing and semantic search optimization. Adept at Responsible AI practices (SHAP, LIME, Fairlearn) ensuring compliance, explainability, and bias detection in production models. Recognized for driving measurable business impact through AI innovation, model optimization, and cross-functional collaboration.

Experience

1. Ai Ml Engineer

   Scale AI

   Sep 2023 — Present · US

   Designed and deployed large language models (LLMs) using PyTorch, Hugging Face Transformers, and ONNX, achieving a 22% improvement in enterprise-grade text generation accuracy.• Engineered real-time data pipelines using Apache Spark, Kafka, and AWS Glue to process 50TB+ ofstructured/unstructured data, reducing ingestion latency by 35%.• Fine-tuned multilingual BERT and GPT models for sentiment analysis and NER, enhancing model precision by 20% across English, Spanish, and Mandarin datasets.• Implemented vector similarity search with FAISS and Pinecone, enabling semantic retrieval at scale and reducing query response time for recommendation systems by 28%.• Developed and optimized computer vision models in TensorFlow and OpenCV for autonomous annotation, improving labeling efficiency by 40% and lowering project costs.• Integrated ML workflows with Kubernetes, Docker, and MLflow, ensuring reproducibility, scalability, and reducing environment-related deployment failures by 15%.• Conducted A/B testing and model monitoring with EvidentlyAI and Prometheus, identifying data drift early and improving model stability in production by 18%.• Built automated feature engineering frameworks in Python, Pandas, and Scikit-learn, accelerating model experimentation cycles by 25% across multiple AI projects.• Collaborated with cross-functional teams to align AI solutions with client needs, delivering scalable ML products that drove a 30% increase in customer satisfaction scores.• Enhanced model explainability with SHAP, LIME, and Captum, providing stakeholders with transparent insights and ensuring compliance with AI ethics and regulatory standards.• Authored technical documentation, design specifications, and knowledge-transfer guides, improving onboarding efficiency for new engineers and reducing team ramp-up time by 20%.

Education

• Harrisburg University of Science and Technology

  Master's degree

• Keshav Memorial institution of commerce and science

  Bachelor's degree, Bachelor of Science