Situation

AI assistants have become the way to use the product and interact with users. The AI assistants are becoming so powerful such that they can take in rich media and do Agentic operations on behalf of the user.
‍
These operational data are not only being massively generated daily, but contain immense business value to various groups of functions. For example, the AI team may want to keep track of the harmfulness of the conversation. The marketing team may want to understand the user's intent from these conversations and promote personalized marketing campaigns.

Problem

To get insights from these data, there are several challenges

1. Visualize the data
2. Search the data
3. Classify the data

Solution

Data generated by How do we make this data searchable?
How do I speed up my unstructured data analysis? How do I visualize a chat session? It's easy to debug a single text or number, but without a chat UI, it's hard to debug a chat conversation with a lot of retrieved context. At Snowflake, we had to build a streamlit chat app to reconstruct the whole chat session and debug it.

A CRUCIAL PRINCIPLE

Richard, drawing from this experience, firmly believes in simplicity as a crucial factor for success.

THE GOAL

With Roe, they aim to uphold this principle by offering an intuitive user experience for data teams and abstracting away the infrastructure.

SIMPLIFIED PROCESS

Users simply load the data and query it—keeping the process straightforward and accessible for all.

A CRUCIAL PRINCIPLE

Richard, drawing from this experience, firmly believes in simplicity as a crucial factor for success.

THE GOAL

With Roe, they aim to uphold this principle by offering an intuitive user experience for data teams and abstracting away the infrastructure.

SIMPLIFIED PROCESS

Users simply load the data and query it—keeping the process straightforward and accessible for all.