Bhargavi T
Senior AI/ML Engineer with 6+ years of experience building production-grade machine learning and Generative AI systems.
About
Senior AI/ML Engineer with 6+ years of experience building production-grade machine learning and Generative AI systems. Focused on LLM engineering, RAG architectures, agentic AI workflows, fine-tuning (LoRA/QLoRA/PEFT), and model evaluation/observability.
Architects scalable MLOps platforms and secure, compliant GenAI deployments across multi-cloud environments (Azure OpenAI, AWS Bedrock, Gemini, Claude APIs). Builds self-hosted low-latency LLM serving, automated drift-aware retraining pipelines, and governance-ready APIs for real-world enterprise use.
- 7 years' experience
- Joined April 2026
Industry experience
- Financial Services
- Healthcare
- Software & Internet
- Professional Services
Work experience· 11 roles
Architected a unified multi-cloud GenAI platform for standardized LLM deployment, evaluation, versioning, and governance; reduced AI rollout time across business units. Designed and deployed enterprise RAG pipelines using LangChain/LlamaIndex, FAISS and custom embeddings with hybrid retrieval and chunking strategies to improve knowledge retrieval speed and reliability. Built agentic assistants using LangGraph/Semantic Kernel and MCP tool orchestration for reasoning, task decomposition, structured extraction, and self-correcting workflows to automate complex analytics. Implemented self-hosted low-latency LLM inference (vLLM/TGI) and reduced external API reliance for cost-sensitive workflows. Added production evaluation and observability using RAGAS and TruLens to monitor retrieval relevance, faithfulness, and answer quality. Developed drift-aware retraining pipelines using MLflow and automation across Azure/AWS, and deployed secure GenAI/RAG APIs using FastAPI, Kubernetes, and Azure Fun
Architected a unified multi-cloud GenAI platform and enterprise RAG pipelines using LangChain, FAISS, GPT-4/Gemini, and hybrid retrieval strategies, achieving ~3x faster knowledge retrieval. Built agentic AI assistants using LangGraph/Semantic Kernel for reasoning, task decomposition, tool use, structured extraction, and self-correcting workflows to automate analytical tasks in finance/operations/risk. Developed LLM summarization/financial Q&A systems that reduced manual analyst effort by ~45%. Implemented LLM evaluation and safety frameworks (toxicity filters, factuality checks, prompt hardening, behavioral tests). Delivered drift-aware retraining pipelines with MLflow and Python automation, and deployed hardened RAG/GenAI APIs using FastAPI + Kubernetes + Azure Functions with RBAC, VNET isolation, TLS, Key Vault/Secrets Manager, and audit logging.
Architected a unified multi-cloud GenAI platform using Python, OpenAI/Azure OpenAI, AWS Lambda, and GCP Vertex AI to standardize LLM deployment, evaluation, and governance, reducing AI rollout time by 40% across business units. Designed and deployed enterprise RAG pipelines with LangChain, FAISS, GPT-4/Gemini, and custom embeddings, delivering ~3x faster knowledge retrieval. Built agentic AI assistants with LangGraph and Semantic Kernel to automate complex analytical tasks in finance, operations, and risk. Developed LLM-powered summarization and Q&A systems to cut manual analyst effort by ~45% and accelerate reporting. Implemented drift-aware retraining pipelines (MLflow, ADF, Lambda) for automated model refresh and drift alerts. Deployed secure GenAI & RAG APIs (FastAPI, Kubernetes, Azure Functions) with RBAC, VNET isolation, TLS, Key Vault, and audit logging. Reduced LLM retrieval latency by 70% with advanced embeddings and Databricks/Spark optimizations. Collaborated with product, a
Education· 6
Hire Bhargavi T today
Post a job and invite Bhargavi, or contact them directly. No hidden commission.