• Integrate cutting-edge AI technologies to solve complex business problems, focusing on vector stores and retrievers while deploying scalable solutions on AWS.
• Design, develop, train, and deploy machine learning and deep learning models to address complex business challenges beyond traditional LLM applications.
• Apply supervised, unsupervised, and reinforcement learning techniques for prediction, classification, recommendation, and optimization use cases.
• Perform hands-on feature engineering, data preprocessing, and model selection to optimize performance and scalability.
• Build and maintain robust experimentation frameworks for model training, validation, and benchmarking.
• Fine-tune open-source and proprietary foundation models using advanced methodologies such as LoRA, PEFT, RLHF, DPO, and instruction tuning.
• Evaluate and refine model performance against various metrics such as accuracy, precision, recall, F1 score, latency, cost, and hallucination rates.
• Implement prompt optimization, model distillation, quantization, and inference acceleration techniques.
• Construct and maintain end-to-end MLOps pipelines supporting the entire lifecycle of model training, deployment, monitoring, and management.
• Implement CI/CD practices for machine learning and LLM workloads using platforms like MLflow, Kubeflow, SageMaker, or Databricks.
• Continuously monitor model drift, data quality, performance, and production reliability.
• Establish automated evaluation and testing frameworks for AI applications.
• Collaborate closely with data scientists and business stakeholders to identify high-impact opportunities.
• Analyze large structured and unstructured datasets using tools such as Python, Pandas, NumPy, Scikit-Learn, PyTorch, or TensorFlow, running experiments, A/B tests, and performance analyses to validate model effectiveness and ensure measurable business outcomes.