2025-10-01 –, Room 1
Writing SQL slows everyone down. Non-technical users can’t, data teams won’t, and leadership waits. While commercial AI-powered tools promise a solution, most are pricey, opaque, and allergic to your reporting requirements. This Ignite talk presents an open-source Text-to-SQL chatbot that prioritizes transparency and user control. It combines advanced prompt engineering & guardrails to reduce hallucinations and ensure generation of reliable SQL queries. It uses an evaluation framework to assess performance by checking syntax accuracy, schema awareness and robustness to ambiguous user inputs. You’ll walk away knowing what works, what breaks, and why building your own AI assistant might just be your smartest move. Query load is not a career path. Offload it to the bot.
Off-the-shelf AI tools are capable of automating text-to-SQL conversion, but many of these solutions are "black boxes" with serious practical limitations like hallucinations, expensive licensing, lack of customizations to company requirements and limited explainability.
By sharing a fully open, explainable approach to Text-to-SQL, this talk is about keeping AI innovation open, collaborative, and community-driven. This session will:
Empower users to build their own AI-powered analytics tools instead of relying on expensive, proprietary software.
Encourage open-source contributions by showcasing how tools like DsPy, LangChain, PostgreSQL, and SQLParse can be combined to create a transparent, community-driven solution.
Foster AI accessibility so that researchers, engineers, and organizations of any size can leverage Generative AI for structured data without compromising on control, security, or cost.
This talk is uniquely relevant to DevOpsDays because it blends:
Self-service analytics (freeing up engineers & data teams from query tickets)
Developer experience (turning LLMs into productive teammates, not flaky oracles)
Open tooling (LangChain, SQLParse, DsPy, Streamlit) to create a transparent system you can fork, debug, and improve without vendor lock-in.
Gamini is a data scientist and AI/ML team lead at Red Hat, Inc. with 7+ years of experience of leading high-visibility advanced analytics projects. Her expertise includes machine learning, statistical forecasting, and building AI-powered automation tools, with a strong focus on practical deployment in business environments. She is a graduate of MS in Business Analytics from Duke University and is currently on track to graduate from MS in Computer Science from Georgia Institute of Technology.