GenAI Engineer - Databricks Mosaic AI

We are seeking for 5+ years experinced AI Engineer with hands-on expertise in Databricks Mosaic AI to design, develop, deploy, and support enterprise-grade Generative AI and Agentic AI solutions. The role will focus on building scalable, secure, and production-ready AI applications leveraging Large Language Models (LLMs), Retrieval-Augmented Generation (RAG), AI Agents, Vector Search, and orchestration frameworks.

The successful candidate will work closely with business stakeholders, data engineers, architects, and AI product teams to translate complex business problems into innovative AI-powered solutions that deliver measurable business value.

Roles & Responsibilities

1. Generative AI & Agentic Application Development

• Build Agentic AI Solutions: Architect, develop, and deploy production-grade Generative AI applications and multi-agent workflows using the Databricks Mosaic AI Agent Framework.
• Vector Search & RAG Architecture: Implement, tune, and scale Retrieval-Augmented Generation (RAG) pipelines leveraging Mosaic AI Vector Search, dense retrieval, and vector databases.
• Orchestration & LLM Invocation: Build agent orchestration logic using LangGraph and LangChain, and utilize AI Query for SQL-based model inferencing.

2. Model Serving & Lifecycle Management (MLOps)

• Model Hosting & Deployment: Deploy, host, and scale fine-tuned LLMs and custom AI models using Databricks Model Serving.
• Governance via Unity Catalog: Manage data pipelines, model registries, fine-tuned weights, and access policies strictly through Databricks Unity Catalog.
• Model Evaluation & Quality Guardrails: Benchmark latency, token usage, precision, and hallucination risks using the Mosaic AI Agent Evaluation Framework and MLflow.

3. Data Pipeline & External Integration

• Data & Spark Pipelines: Partner with data engineering teams to construct scalable Apache Spark (PySpark) and SQL pipelines for continuous vector embedding and index generation.
• API & Model Hub Integration: Integrate external model hubs and SDKs (including OpenAI SDK and Hugging Face) with core enterprise platforms.