Databricks GenAI Solutions Architect

Job description:

We are seeking a highly skilled Databricks GenAI Solutions Architect to lead the design, development, and deployment of enterprise-grade Generative AI solutions on the Databricks Lakehouse Platform. This role combines deep expertise in Large Language Models (LLMs) with comprehensive knowledge of the Databricks ecosystem, including Mosaic AI, MLflow, Unity Catalog, and Delta Lake. You will be responsible for architecting scalable LLM-based applications, implementing RAG pipelines, establishing MLOps/LLMOps best practices, and mentoring the team on cutting-edge GenAI technologies within the Databricks environment.

Responsibilities:

• Architect end-to-end GenAI solutions including RAG (Retrieval-Augmented Generation), AI agents, and fine-tuned models on Databricks.
• Evaluate and select appropriate LLMs (OpenAI, Anthropic Claude, Azure OpenAI, DBRX, Llama, Mistral) based on use case requirements and cost optimization.
• Design and implement scalable vector search solutions using Databricks Vector Search and Mosaic AI.
• Establish prompt engineering best practices, optimization techniques, and prompt management strategies.
• Implement compound AI systems and multi-agent architectures using Databricks Agent Framework.
• Design and implement GenAI solutions leveraging Databricks Mosaic AI capabilities including Model Serving, Foundation Model APIs, Vector Search, AI Gateway, and Feature & Function Serving.
• Architect data pipelines using Delta Lake and Spark for preprocessing, chunking, and embedding generation at scale.
• Implement robust data governance using Unity Catalog for model assets, features, and vector indices.
• Design and manage MLflow experiments for LLM evaluation, tracking, and model registry.
• Design LLMOps pipelines for evaluation, monitoring, A/B testing, and continuous improvement.
• Implement guardrails, safety measures, content filtering, and responsible AI practices.
• Build observability solutions for LLM applications including latency monitoring, cost tracking, and quality metrics.
• Establish CI/CD pipelines for GenAI model deployment using Databricks Workflows and Asset Bundles.
• Lead technical decisions and conduct architecture reviews for GenAI initiatives.
• Collaborate with business stakeholders to translate requirements into scalable technical solutions.
• Mentor engineers and data scientists on GenAI development standards and Databricks best practices.
• Create technical documentation, reference architectures, and reusable solution accelerators.

Qualifications:

• 5+ years of relevant experience in ML/AI with 3+ years focused on GenAI/LLM applications.
• Deep hands-on experience with LLM APIs (OpenAI, Anthropic Claude, Azure OpenAI, Google Vertex AI).
• Strong RAG implementation experience including document processing, chunking strategies, embedding models, and retrieval strategies (hybrid search, re-ranking, query transformation).
• Experience with vector databases (Pinecone, Weaviate, ChromaDB, FAISS, Azure AI Search, Databricks Vector Search).
• Advanced prompt engineering skills including chain-of-thought, few-shot learning, and prompt optimization.
• Proficiency with LLM orchestration frameworks (LangChain, LangGraph, Semantic Kernel).
• Understanding of fine-tuning techniques (LoRA, QLoRA, PEFT, full fine-tuning).
• 3+ years hands-on experience with Databricks platform.
• Strong proficiency in PySpark and Spark SQL for large-scale data processing.
• Experience with Databricks Mosaic AI suite: Model Serving, Vector Search, Foundation Model APIs, AI Gateway.
• Expertise in MLflow for experiment tracking, model registry, and deployment.
• Working knowledge of Unity Catalog for data and AI governance.
• Experience with Delta Lake architecture and optimization.
• Familiarity with Databricks Workflows for orchestration.
• Strong programming skills in Python; familiarity with SQL.
• Production deployment experience for LLM applications at scale.
• Excellent system design and distributed architecture skills.
• Understanding of API design, microservices, and cloud-native patterns.
• Experience with containerization (Docker, Kubernetes) for model serving.

Skills

Databricks, Delta Lake, LLM, LangChain, MLflow, Mosaic AI, OpenAI, PySpark, RAG, Unity Catalog