Hey Product Hunt! I’m Max, founder & CEO of Basedash. We first launched @Basedash earlier this year, and now we're back with our first big update.
Today we’re launching the Basedash Agent: an AI chat agent that can answer questions, create visualizations, and surface insights on all your company's data.
The idea is simple: you ask a question in chat, and the agent does the heavy lifting—running SQL on your database, pulling context from 600+ SaaS tools (Linear, HubSpot, GitHub, Slack, and more), and returning a clear answer. When it helps, it can generate a longer report including charts and visualizations that you can pin to dashboards.
You can also add custom context about how your company works, the terminology you use, and the metrics your team cares about. The Basedash Agent takes all this into account every time you ask it a question.
There’s also a Slack integration: just mention Basedash in any channel and get answers right in Slack.
Why it’s different - Grounded in your real, live data - Answers can be text, reports, or visualizations—not just a snippet - Built for teams: now anyone can chat with your business data
Ask it things like - Which segments are trending up this week? Visualize it. - List recent signups that best fit our ICP and explain why. - What’s holding back activation, and which experiments should we try? - What patterns do high-conversion workspaces share?
We’d love your feedback! Tell us the hardest question you wish your team could answer—during launch day we’ll run real prompts and share results.
@maxmusing @long_hoang14 thank you so much, that's very kind of you! Excited to hear what you think.
Report
@maxmusing The operation is simple and does not require any technical skills. You can use natural language to create the chart you want, and AI automatically generates visual content. The effect is great
@maxmusing @frank_yang_tao Totally agree!! that’s the magic of it. If you can describe what you need, the system handles the rest. Our goal was to make analytics feel as simple as having a conversation. Appreciate you trying it out!
We don't just rely on the models themselves. There's a ton of context engineering and optimizations baked into every AI call in Basedash - which we test and validate thoroughly with our eval system (which is constantly growing as we run into crazy use cases)
We use a variety of techniques like: - RAG - algorithmic schema context building - conversation summarization - learning from existing queries / charts - fine-grained tools - contextual prompt engineering - workflows / agentic loops / guardrails - etc etc
Our goal is to continue adding and building on these techniques while also staying on the latest AI capabilities. For instance, we upgraded some of our models to GPT-5 within hours of it launching!
@bryan_hunter Couldn’t agree more. The model is just one piece, the real magic is all the context engineering, evals, and tooling around it. That’s what makes the system feel reliable and actually usable day-to-day. Huge shoutout to Bryan for leading so much of that work.
Very excited about this launch! We’ve been using Basedash Agent ourselves and the not so secret truth is that we basically run our entire business of Basedash. Looking for better insights into activation? Basedash gives us those. Looking for bugs that harm user experience? Basedash gives us those. It’s like having a world class data analyst on your team and we’re so excited for you to try it.
I use Basedash Agent almost every day for helping debug customer and internal issues. We have @github, @Sentry and our own application data (like queues and crons) available in our warehouse. Simply asking `@Basedash` something like "hey some users are experiencing slower AI chat responses suddenly… anything stand out to you?" And it can produce a really helpful report to help jump start a fix in most cases.
That's just one way I like to use this feature, but there are so many more that help automate some of the day-to-day knowledge sharing like:
- Standups/status updates about what's shipping based on @Linear and @github data - Analyzing user behavior and producing daily insights using our application data in @PostgreSQL - Turning chats in to charts that can be shared to Slack
When you can get your data together in a warehouse and avoid all the ad-hoc MCP glue and address common issues with LLMs like hallucinations and inefficient workflows it truly unlocks a next-level experience of collaboratively chatting with your data.
Thanks @ryangilbert! We've always thought of Basedash as AI-native, so building a centralized data agent just made sense. It's crazy to think that we didn't have this at launch.
This launch has been a long time coming, and we're really excited for people to use it.
Since our initial launch of Basedash, we've spent nearly all our time adding functionality, reliability, performance, and tool integrations to this AI agent. It wasn't until GPT-5 arrived that we finally had the capabilities we'd always wanted, namely, the ability to reliably generate multiple charts and cross-reference data already created within an existing workspace.
With this agent, you can:
Generate dozens of visualizations and different perspectives on your data with a single prompt, then add them to dashboards in seconds.
Give open-ended prompts for the Agent to dig into your data in ways that were previously impossible, even for the most skilled data analysts. For example:
"What interesting people have signed up to our product recently?"
"What features aren't we using?"
"How's our website performing versus our competitors or industry averages?"
etc
These capabilities make it possible for non-technical teammates to dig into data on their own. Now, anyone can explore the data without worrying about how to visualize it or what SQL to write, but simply focusing on the questions they want answered.
The collaborative aspect is especially important for data teams and those with a security mindset. This isn't just a standalone AI chatbot use in isolation, where you have to manage all the context, access, and schema. It's a shared workspace where data teams can configure access, restrict permissions, and ensure that people interacting with the data only see what they need or should have access to. Teams can also review insights, work together, and make sure the AI agent isn't giving misleading signals.
Plus, with the agent working inside Slack, the barrier to entry is basically zero. Marketing, sales, product, engineering, and support teams can all start asking Basedash for insights, find trends, dig into bugs, and understand conversion rates right inside Slack. What used to take internal data teams or consultants days or weeks now takes minutes.
We're also seeing agencies use Basedash for their clients: setting up permissions, creating visualizations, and delivering valuable insights at a speed that makes their team feel 10x bigger.
The capabilities this launch brings are unique in the industry, and we're excited for you to see what's next. We're only just getting started.
Have been super excited about this launch. Being able to answer complex questions within the course of a meeting, without jumping into the context of SQL, lets my team fly. The evals make it that much easier to justify using.
Congrats guys! I've been following what you're doing for a while and continue to be impressed with your taste and creativity. We'll be looking to integrate soon :)
Appreciate it @aidanhornsby! Happy to help you get set up when you're ready, just reach out.
Report
Data analysis by AI agents has become a trend, and I believe many people will use Basedash. Would you like to know if Basedash can connect to CRM data sources? This will be very convenient for intelligently analyzing sales data!
Thanks @wayne_appgrowing! We can connect to most CRMs including HubSpot, Salesforce, Zoho so you can ask questions about your pipeline, funnel, sales, etc.
Replies
Basedash
Hey Product Hunt! I’m Max, founder & CEO of Basedash. We first launched @Basedash earlier this year, and now we're back with our first big update.
Today we’re launching the Basedash Agent: an AI chat agent that can answer questions, create visualizations, and surface insights on all your company's data.
The idea is simple: you ask a question in chat, and the agent does the heavy lifting—running SQL on your database, pulling context from 600+ SaaS tools (Linear, HubSpot, GitHub, Slack, and more), and returning a clear answer. When it helps, it can generate a longer report including charts and visualizations that you can pin to dashboards.
You can also add custom context about how your company works, the terminology you use, and the metrics your team cares about. The Basedash Agent takes all this into account every time you ask it a question.
There’s also a Slack integration: just mention Basedash in any channel and get answers right in Slack.
Why it’s different
- Grounded in your real, live data
- Answers can be text, reports, or visualizations—not just a snippet
- Built for teams: now anyone can chat with your business data
Ask it things like
- Which segments are trending up this week? Visualize it.
- List recent signups that best fit our ICP and explain why.
- What’s holding back activation, and which experiments should we try?
- What patterns do high-conversion workspaces share?
We’d love your feedback! Tell us the hardest question you wish your team could answer—during launch day we’ll run real prompts and share results.
@maxmusing all the best for your launch!
Basedash
@maxmusing @long_hoang14 thank you so much, that's very kind of you! Excited to hear what you think.
@maxmusing The operation is simple and does not require any technical skills. You can use natural language to create the chart you want, and AI automatically generates visual content. The effect is great
Basedash
@maxmusing @frank_yang_tao Totally agree!! that’s the magic of it. If you can describe what you need, the system handles the rest. Our goal was to make analytics feel as simple as having a conversation. Appreciate you trying it out!
Basedash
We don't just rely on the models themselves. There's a ton of context engineering and optimizations baked into every AI call in Basedash - which we test and validate thoroughly with our eval system (which is constantly growing as we run into crazy use cases)
We use a variety of techniques like:
- RAG
- algorithmic schema context building
- conversation summarization
- learning from existing queries / charts
- fine-grained tools
- contextual prompt engineering
- workflows / agentic loops / guardrails
- etc etc
Our goal is to continue adding and building on these techniques while also staying on the latest AI capabilities. For instance, we upgraded some of our models to GPT-5 within hours of it launching!
Basedash
@bryan_hunter Awesome engineering work on this!
Basedash
@bryan_hunter hugely important engineering work, well done sir!
Basedash
@bryan_hunter Couldn’t agree more. The model is just one piece, the real magic is all the context engineering, evals, and tooling around it. That’s what makes the system feel reliable and actually usable day-to-day. Huge shoutout to Bryan for leading so much of that work.
Basedash
Very excited about this launch! We’ve been using Basedash Agent ourselves and the not so secret truth is that we basically run our entire business of Basedash. Looking for better insights into activation? Basedash gives us those. Looking for bugs that harm user experience? Basedash gives us those. It’s like having a world class data analyst on your team and we’re so excited for you to try it.
Basedash
I use Basedash Agent almost every day for helping debug customer and internal issues. We have @github, @Sentry and our own application data (like queues and crons) available in our warehouse. Simply asking `@Basedash` something like "hey some users are experiencing slower AI chat responses suddenly… anything stand out to you?" And it can produce a really helpful report to help jump start a fix in most cases.
That's just one way I like to use this feature, but there are so many more that help automate some of the day-to-day knowledge sharing like:
- Standups/status updates about what's shipping based on @Linear and @github data
- Analyzing user behavior and producing daily insights using our application data in @PostgreSQL
- Turning chats in to charts that can be shared to Slack
When you can get your data together in a warehouse and avoid all the ad-hoc MCP glue and address common issues with LLMs like hallucinations and inefficient workflows it truly unlocks a next-level experience of collaboratively chatting with your data.
One of our most exciting launches so far!
Basedash
@github @Basedash @drk we wouldn't be here without you!!!
Workspaces
Love this approach! How long did you and the team go back and forth when deciding if this was the next move?
Basedash
Thanks @ryangilbert! We've always thought of Basedash as AI-native, so building a centralized data agent just made sense. It's crazy to think that we didn't have this at launch.
Basedash
@ryangilbert thanks so much for the kind words, Ryan!
Basedash
This launch has been a long time coming, and we're really excited for people to use it.
Since our initial launch of Basedash, we've spent nearly all our time adding functionality, reliability, performance, and tool integrations to this AI agent. It wasn't until GPT-5 arrived that we finally had the capabilities we'd always wanted, namely, the ability to reliably generate multiple charts and cross-reference data already created within an existing workspace.
With this agent, you can:
Generate dozens of visualizations and different perspectives on your data with a single prompt, then add them to dashboards in seconds.
Give open-ended prompts for the Agent to dig into your data in ways that were previously impossible, even for the most skilled data analysts. For example:
"What interesting people have signed up to our product recently?"
"What features aren't we using?"
"How's our website performing versus our competitors or industry averages?"
etc
These capabilities make it possible for non-technical teammates to dig into data on their own. Now, anyone can explore the data without worrying about how to visualize it or what SQL to write, but simply focusing on the questions they want answered.
The collaborative aspect is especially important for data teams and those with a security mindset. This isn't just a standalone AI chatbot use in isolation, where you have to manage all the context, access, and schema. It's a shared workspace where data teams can configure access, restrict permissions, and ensure that people interacting with the data only see what they need or should have access to. Teams can also review insights, work together, and make sure the AI agent isn't giving misleading signals.
Plus, with the agent working inside Slack, the barrier to entry is basically zero. Marketing, sales, product, engineering, and support teams can all start asking Basedash for insights, find trends, dig into bugs, and understand conversion rates right inside Slack. What used to take internal data teams or consultants days or weeks now takes minutes.
We're also seeing agencies use Basedash for their clients: setting up permissions, creating visualizations, and delivering valuable insights at a speed that makes their team feel 10x bigger.
The capabilities this launch brings are unique in the industry, and we're excited for you to see what's next. We're only just getting started.
Basedash
@tomjohndesign so so excited!
Have been super excited about this launch. Being able to answer complex questions within the course of a meeting, without jumping into the context of SQL, lets my team fly. The evals make it that much easier to justify using.
Basedash
Thanks @ted_spare! Love watching our evals go up with every update.
Optimist
been using basedash at composio for a bit and it’s great for getting quick answers from data
Basedash
@cryogenicplanet thanks so much! we're so so lucky to work with amazing companies like composio :)
Basedash
@cryogenicplanet love to hear it, working to make Basedash better every day
Layercode
Congrats guys! I've been following what you're doing for a while and continue to be impressed with your taste and creativity. We'll be looking to integrate soon :)
Also, loved the launch video!
Basedash
Appreciate it @aidanhornsby! Happy to help you get set up when you're ready, just reach out.
Data analysis by AI agents has become a trend, and I believe many people will use Basedash. Would you like to know if Basedash can connect to CRM data sources? This will be very convenient for intelligently analyzing sales data!
Basedash
Thanks @wayne_appgrowing! We can connect to most CRMs including HubSpot, Salesforce, Zoho so you can ask questions about your pipeline, funnel, sales, etc.