How to debug AI chatbots with Preswald
Learn how to use Preswald to analyze and visualize chatbot logs. This guide covers deployment, data processing, and insights with easy-to-use tools
Learn how to use Preswald to analyze and visualize chatbot logs. This guide covers deployment, data processing, and insights with easy-to-use tools
Analytics will no longer be an isolated function. It will be the operating system of the modern organization, powering decisions, automating processes, and predicting the future.
With a code-first approach, you can automate the entire setup process and start running analytics directly on structured data with minimal effort.
“Make something people want” has long been the mantra of startups. But what happens when humans aren’t the ones making decisions anymore? The analytics tools we’ve built are fundamentally broken for this new world.
Run Python directly in the browser with Pyodide. This makes sharing code easier, and opens up new ways to build interactive, web-based data apps.
If you deploy a containerized app, it behaves the same on your laptop as it does in production. So why can’t our analytics dashboards do the same?
CSV files seem simple. Just plain text, right? But when they get large, they are not great formats for data exploration. CSVs lack indexing, compression, and structured access.
If you could build data apps at the speed of thought—what would happen? When building data apps becomes this fast, this cheap, and this effortless.
Sometimes the shiny, purpose-built solution isn’t actually better. Sometimes, what we need isn’t another feature-rich gadget, but something simpler, more versatile, and less likely to create chaos down the line.
The problem with most analytics projects is that they start with the wrong question. People say, “We need a dashboard,” but that’s like saying, “We need a hammer,” without knowing what you’re building.
Have you ever inherited what can only be described as the Franken-stack of analytics? The total cost— is something much harder to see.
How Ibis simplifies analytics across multi-backend systems with a unified Python API. Ibis offers a DataFrame-like API, by translating Python operations into backend-specific queries.
Sure that chat-with-your-data app was cute in 2022. If you’re handing over the keys to an LLM without interrogating its output, you’re outsourcing the very thing that makes analytics valuable: human curiosity and creativity.
If you’re new to Parquet, think of it as a data format optimized for analytics. Unlike a traditional spreadsheet or a database table that stores data row-by-row, Parquet stores data column-by-column.
The rise of AI agents as SDK consumers will fundamentally change how we should design developer tools. Prompt-based interfaces are becoming an important part of the developer product experience.
From earthquakes to action: Real-time tools for tsunami preparedness. To address this need, we quickly built a custom data app using the USGS Earthquake Hazards Program API and Preswald