Livedocs - The general data agent
by•
Livedocs is a general data agent that can do BI, ML, build dashboards, and write code and queries (anything a human data scientist would do!)
Livedocs helps you understand your data instantly. Upload a CSV, spreadsheet, or connect your database, then ask questions in plain English. Livedocs uses AI to generate charts, metrics, and clear answers—no SQL, no dashboards, no setup. Built for anyone who needs insights fast, whether you’re technical or not. Just bring your data and start asking.
Replies
Amazing product! Congrats to the Livedocs team!
LiveDocs
@natiakourdadze Thanks Natia, appreciate your support!
Link Texting
On prem model wins. ;) Nicely done.
LiveDocs
@datarade Thanks! You can run Livedocs locally, on-prem, on a hosted cloud or even your own cluster!
Congratulations on the launch! Really like the overall interface and the quality of the analysis. Also had a couple of questions, does the system auto-detect relationships between data sets or do they have to be added to the context tab? And how are null values handled across different models?
LiveDocs
@mustassim Hi Syed, thanks for your support! Yes, Livedocs will auto-detect the schema of your databases. For other datasets like files, etc, too it would search across available data for anything that helps it get the full picture.
Livedocs also cleans up the data and looks for missing/null values. You can pick any model you like, so larger, more recent models are better than smaller models at cleaning and analyzing data.
Curious to learn about what kind of data you are analyzing with Livedocs
Agnes AI
nice data tool - one thing which is crucial is the accuracy rate. How is the average accuracy rate of LiveDocs?
LiveDocs
@cruise_chen Hi Cruise! Accuracy was a huge priority for us while building Livedocs, which is why Livedocs always shows the source code and queries used and explains how it got the results. We're also performing in the top 10 on some kaggle competitions, so that's a great sign
Huge congrats on the update and launch @arslnb !
Early days can be wild with numbers are you finding your conversions line up with expectations so far, or does it feel like something isn’t clicking the way you hoped?
That’s often where early traction decisions matter most. :)
This makes life much easier for data scientists 👍👍
LiveDocs
@maria_doliashvili Thanks, Maria!
From my experience, the biggest bottleneck with data tools isn’t lack of power — it’s the friction between questions and answers. Most teams don’t want to build dashboards or write SQL just to explore an idea.
I like how LiveDocs keeps the workflow conversational but still allows you to go deeper with notebooks when needed. That flexibility is important — especially for founders and ops teams who want quick clarity first, then depth.
Curious how teams handle versioning or sharing insights across stakeholders once analyses evolve. Overall, this feels very thoughtfully designed. Nice work 👏
Has been great so far. I love being able to collaborate on a notebook with my colleagues to build meaningful reports for our clients. Never thought that doing the analysis part with “messy” data would be so seamless.
Our work takes us to remote locations in Australia with limited internet connectivity. How easy is it to set up offline?
SigniFi
Love that you didn't dumb it down too much—keeping the full notebook mode with SQL/Python is a smart move for power users. 🐍 regarding the environment: Can I import standard Python libraries (like Scikit-learn or Prophet) for forecasting, or is the sandbox restricted to basic Pandas operations?
LiveDocs
@yoang_loo Thanks! And yes — you can pip install pretty much any Python library you want (scikit-learn, Prophet, etc.). Livedocs notebooks are essentially a superset of Jupyter, so if it works in Jupyter, it works here too.
Thanks @arslnb! Great tool that makes my data analysis easier and more fun! Just curious if we would enable multiple language support sometime in the future?