Data Analyst Agent

💡

Built right into Sapience and available immediately “out of the box” is the Sapience Data Analyst Agent. This uses the Sapience Coding Agent tool behind the scenes, and is perfect for analyzing data files, generating charts and visuals and performing sophisticated data analysis and transformation.

How to use it:

Click the Agent Selector;

Click Browse all Agents, and look for the Sapience Data Analyst Agent

Should look like this:

*The Data Analyst Agent is a global agent that comes pre-configured for BI and analysis tasks.*

It will take you to the main Chat interface;

Add tabular data to the conversation;

Ask it to generate an image or chart for you (see more below).

Sapience Coding Agent for Visualization & Data Science

Intelligent data-file detection: When you upload a data file, or tabular data, which usually means .xls, .xlsx, .csv or .json, Sapience will detect this and prompt you to determine if you want to use the Sapience Coding Agent, or if you want to deal with it like a regular document.

Example: In the screenshot below you can see that the user has added an Excel file, and Sapience then prompts asking how you want to treat the file. The coding agent spins up a container and can perform sophisticated analysis or image generation. Ask it to generate static charts or interactive dashboards and it will. Just be sure to be specific in what you want it to generate.

Adding tabular data to Agents that support using the Sapience Coding Agent as a tool will cause Sapience to ask you if you want to use the coding agent for analysis versus treating it as a regular document.

How to try it:

start a chat with a real agent from your org or that you’ve created. If you don’t have any, use the Sapience Agent;

upload tabular data, which means a file ending in xslx, xls, CSV, TSV, json.

Select Yes, you want to use the Coding Agent

Describe the file: important! putting in a bad description will hamper results.

Once you return to the chat, ask the AI about the file. Tip: if you ask the AI to explore the data first you will get better results.

Static visual: ask it to create a static visual.

Example prompt: “Create a static image/chart showing the data from the 1st column in the dataset on the Y access. Give it to me as a histogram to start with. Use values from the ‘revenue’ column”.

Interactive app: Sapience can even code up an interactive dashboard or module for you, and write all of the HTML + CSS + Javascript necessary to do so! Use the key phrase “interactive visual” or “interactive dashboard”.

Example prompt: “Create an interactive visual, being a 3D model that I can rotate. For the X axis use {data} for the Y axis use {data2} and the Z axis use {data3}. Use small dots for the data points as there are many rows in the excel.”

Real Life Examples

Here are some useful examples:

This is a real-life Dashboard created by the Sapience Coding Agent, using a dataset that had about 1000 rows and 250 columns (so 250,000 datapoints). You can’t tell from this screenshot, but this Dashboard is dynamic, and you can mouse over the different items and it then displayed additional data on mouse-over. This would take a human a full work week to build!

Make sure you check the undertext of the Agent replies to see files it may have generated while the Coding Agent was working. In this screenshot, you can see that it created a powerpoint, as well as an excel file and several charts.