Snowflake recently announced Cortex, their LLM for unstructured data/questions/copilot/assistant. I was at Snowflake Summit earlier this month and came across a lot of AI tools for data teams similar to Cortex, like Secoda, Glean, Gemini, dbt’s AI and a bunch more. I want to know how people are actually using AI in their data workflow.
Has anyone implemented AI for their data/analytics teams? What tools are you using? Where in your workflows are you using AI? Is this all hype??
Avatar I haven’t fully implemented AI in my data workflow yet, but I’ve been exploring it a lot recently, especially after hearing about tools like Snowflake Cortex, Secoda, and dbt’s AI features. One thing I’ve noticed is that many of these tools are trying to make it easier for non-technical users to interact with data—like writing SQL queries just by typing a question in plain English or generating documentation automatically. I’m still figuring out how effective they are in real-world use. Some seem promising for speeding up routine tasks like exploring datasets or cleaning data, but I wonder how reliable they are for more complex analysis. I’ve also been trying to understand where exactly these tools fit in—should they be used during data exploration, dashboarding, or governance? It does feel like there’s a lot of hype right now, so I’m curious to know from others who are ahead in this journey: which tools have actually made a difference in your workflow? And are there areas where AI still falls short?
AI is increasingly finding real utility in modern data workflows, particularly in tasks that require automation, context understanding, or faster iteration. Tools like Snowflake Cortex, Secoda, and Glean are being used to streamline data discovery, documentation, and querying. For instance, many data teams use AI copilots to auto-generate SQL queries from natural language prompts, reducing dependency on technical roles for basic analysis. In tools like dbt, AI is assisting with documentation, code suggestions, and even identifying anomalies or lineage issues proactively. Some teams also integrate LLMs to answer internal data questions, generate dashboards, or summarize large datasets into business-friendly insights. While there’s definitely some hype, the practical use cases—like speeding up data exploration, simplifying governance, and enabling self-serve analytics—are proving valuable. Adoption depends heavily on the maturity of the data stack and the team’s openness to experimentation, but many are already seeing time and cost efficiencies, especially in early-stage analysis and reporting.
