Senior AI Engineer

Implement multi-agent intelligence frameworks (LangGraph, CrewAI, AutoGen) to enable reasoning, coordination, and adaptive decision-making across specialized AI agents.
• Design and operationalize multi-modal AI pipelines combining text, image, tabular, and graph data using transformer-based architectures (BERT, CLIP, LLaVA, T5, Whisper, etc.) for unified intelligence.
• Build scalable RAG and Graph-RAG systems integrating vector stores and knowledge graphs (Neo4j, AWS Neptune) to enable contextual retrieval, semantic linking, and entity-aware reasoning.
• Develop and productionize transformer-based models for NLP, vision-language understanding, and sequential prediction tasks leveraging Hugging Face, PyTorch, and TensorFlow ecosystems.
• Implement advanced Python-based backend services for inference orchestration, async job handling, and distributed data workflows supporting high-throughput AI operations.
• Establish end-to-end LLMOps and MLOps pipelines on Databricks (AWS) integrating MLflow, feature stores, model lineage, prompt evaluation, and continuous retraining frameworks.
• Apply traditional AI/ML and statistical modeling techniques (regression, clustering, forecasting, ensemble methods) alongside deep learning models for hybrid interpretability and explainability.
• Engineer state and memory management subsystems that preserve context, track embeddings, and enable agents to reason temporally across multiple modalities and interactions.
• Implement Responsible AI practices—bias detection, explainability dashboards, data ethics checks, and performance governance ensuring fairness and transparency of deployed models.
• Continuously research, benchmark, and productionize innovations in multimodal transformers, generative modeling, and agentic orchestration to drive enterprise-scale intelligence and automation.