AI Agents on Snowflake for Data-Driven Decision Making by Sarah Lina LachhebAI Agents on Snowflake for Data-Driven Decision Making by Sarah Lina Lachheb

AI Agents on Snowflake for Data-Driven Decision Making

Portfolio project
Sarah Lina Lachheb
I designed and developed AI agents on Snowflake (for a media company) to enable users to interact with enterprise data through natural language (finance agent, user 360 agent, gaming agent, dynamic paywall agent and more..)
The goal was to move beyond traditional dashboards and allow business users to ask questions, explore KPIs and retrieve insights without having to write SQL or navigate complex reporting environments.
The solution combined Snowflake's data platform, semantic models and AI capabilities to create agents grounded in trusted business data.
What I worked on:
Designed the agent architecture and interaction logic
Connected agents to governed Snowflake data
Built semantic layers to provide business context and ensure consistent answers
Defined the questions and analytical use cases the agents should handle
Optimized prompts and agent instructions for accuracy and relevance
Considered token consumption and cost efficiency when designing agent workflows
Tested responses against expected business answers and edge cases
Designed the experience around real business use cases rather than generic AI interactions
The result was a conversational data experience where users could interact with their company's data in a much more natural way.

0 likes · 0 views · Nov 15, 2026

I designed and developed AI agents directly within Snowflake, combining enterprise data, semantic models and LLM-powered reasoning to make complex data accessib

Tags