Ai Engineer

We are looking for a hands-on AI Engineer to help extend our existing data product with a semantic and conversational AI layer. This role will focus on building a robust semantic layer on top of our established domain-driven data model, and on designing AI-powered user experiences — conversational agents, chatbots, and Streamlit-based applications that let users interact naturally with the underlying data.

Key Responsibilities

• Design and build a semantic layer on top of our existing domain-driven data model.
• Translate physical data structures and domain entities into business-friendly concepts that can be reliably consumed by AI applications.
• Define and manage business entities and relationships, measures/metrics/KPIs, terminology and synonyms, dimensions and hierarchies, and business rules.
• Build semantic models that improve the accuracy and consistency of natural-language-to-data interactions.
• Develop and deploy AI agents and conversational applications that let users query and interact with product data using natural language.
• Build user-facing AI applications using Streamlit, Snowflake-native capabilities, or appropriate third-party agent/chatbot frameworks.
• Develop agent workflows that interpret user intent, query structured data, apply business context, and present useful responses.
• Integrate AI applications with existing product data, APIs, and services.
• Develop and maintain prompt strategies, system instructions, tool definitions, and guardrails.
• Test and evaluate AI responses for accuracy, consistency, and adherence to business logic.

Qualifications:

• 2 – 3 years with strong experience building and deploying LLM-powered applications and AI agents.
• Strong Python and SQL skills, with experience working with complex relational data models.
• Experience building semantic models, semantic layers, or business metadata layers for analytical or AI use cases.
• Understanding of agentic architecture: tool/function calling, agent orchestration, context management, structured outputs, and multi-step workflows.
• Understanding of RAG, embeddings, vector search, and grounding techniques where appropriate.
• Hands-on Snowflake experience strongly preferred (Cortex AI, Cortex Analyst, semantic models/views, Cortex Search, Streamlit in Snowflake).
• Experience with agent frameworks (LangGraph, LangChain, Semantic Kernel, AutoGen) and enterprise LLM platforms (OpenAI, Azure OpenAI, Anthropic, AWS Bedrock) is a plus.
• Experience with AI evaluation, observability, and response-quality testing is valued.
• Applied AI engineer with strong data modeling instincts — able to bridge an established domain-driven data model with modern AI applications, rather than an AI researcher focused on training models.