DecisionBox

Autonomous AI Discovery For Your Data — Open Source

49 followers

Autonomous AI Discovery For Your Data — Open Source

49 followers

DecisionBox is an open-source project and connects to your data warehouse, runs autonomous AI agents that write and execute SQL, and surfaces validated insights and actionable recommendations, without you asking a single question.

This is the 2nd launch from DecisionBox. View more

DecisionBox - Ask

Launching today

Your team's findings, in plain English

DecisionBox agents ran discoveries and found 130+ insights last week. Now the rest of your team needs to find the right one. Ask is plain-English search over every insight and recommendation discovery produced. Open source, your infra, no hallucinations.

Free

Launch tags:Open Source•Artificial Intelligence•GitHub

Launch Team

Tempo - Align Work to Strategy — Turn Jira Data Into Strategic Clarity

Turn Jira Data Into Strategic Clarity

Promoted

Maker

📌

Hey Product Hunt 👋 We launched DecisionBox here last week. Thank you to everyone who upvoted, tried it, broke it, filed issues. Today we're back with a feature we've been sitting on for a bit: **Ask**. Quick context in case you missed last week: DecisionBox runs autonomous AI agents on your data warehouse. Point it at BigQuery, Snowflake, Redshift, Databricks whatever — pick a domain pack for your industry — and it goes off and actually finds stuff. Real stuff. Validated against your data. Not hallucinated. At the end of a run you get a pile of insights and recommendations. Here's the problem that showed up almost immediately. The runs work. They find things. But the dashboards start filling up fast. We ran a discovery against a vacation rental platform and it produced 130+ insights from a single run. A few days later one of our test users asked me "hey, didn't you find something about cancellation rates?" — and I had to go dig through three different discovery runs to find the right one. That's the problem Ask solves. First thing I want to say: it is not a chatbot. I know how "AI" + "ask a question" reads in 2026. It is not that. Ask doesn't generate answers. It retrieves them. You type a question in plain English — *what are our biggest retention risks?*, *which recommendations should we prioritize?* — and it surfaces the actual validated insights and recommendations your agents already produced. With the run ID, the confidence score, the affected count, the source query. Traceable all the way back to the SQL. The semantic part is the thing I actually obsessed over. Ask for "churn signals" and it pulls up findings about cancellation rate, repeat-booking, guest satisfaction — even if none of those exact words are in your question. The whole point is you shouldn't need to remember the phrasing of a finding to get back to it. One more thing worth flagging: still open source, still AGPL v3, still runs in your own infra. This is a community release, not a paywall feature. (Our enterprise tier adds SSO/RBAC/audit — it doesn't gate Ask.) What I'm actually curious to know: if you've ever stared at a dashboard thinking *"we have too many insights and nobody looks at them"* — does this help? That was the pain I built it for. I'll be hanging around in the comments all day. Ask me literally anything. — Can (DecisionBox team)

Report

3d ago

Maker

Here is a blog post about DecisionBox Ask feature, we have used WanderBricks sample database on Databricks: https://decisionbox.io/blog/ask/

Report

6h ago

Previous DecisionBox Launches

DecisionBoxAutonomous AI Discovery For Your Data — Open Source

Launched on April 7th, 2026

Forum Threads

p/decisionbox

•

9d ago

We pointed DecisionBox at a data set it had never seen. Here's what it found under an hour.

Most AI analytics tools need you to ask the right questions. We wanted to see what happens when you don't ask anything at all.

We took a public Kaggle dataset, a 109 million raw e-commerce dataset from a multi-category store. No documentation, no data dictionary, no context. We pointed DecisionBox at it and hit "Start."

p/decisionbox

•

9d ago

DecisionBox AI Agent ran 233 SQL queries on Databricks in 76 minutes. Here's what it found.

Hey Product Hunt!

I'm Can, co-founder of DecisionBox. I wanted to share a real case study of what our open-source AI agent does when you point it at a data warehouse.

View all