Coda AI with Janvi Kalra
Learning from a great example
June 13, Coda launched their AI touch in public beta. and Aug 15, the 4the Coda Ai element was added: AI chat.
I had the opportunity to test the AI (chat) and understood the opportunities it presents and also some limitations we’d better be aware of. I’ll focus on both in this blog. My main message is that you should try to reproduce the work she shows.
We take as example a great and inspiring showcase worthwhile watching.
The example shows a few intriguing applications of AI. Please note that this example was made before the introduction of Coda chat on August 15:
- the AI created a table with 10 to visit places
- the AI created a second table with restaurants including budget and website
- the AI generated a day planning for 3 days based on these two tables
- the AI content was linked to a pack to show places on a map
We thus have two tables filled out with content the AI grabbed and we should be aware that this data may already be outdated. The LLM feeding Coda is not using real time internet data, it is about 2 years old. Real time internet data is something makers are asking for since it opens up so many great use case, but until today, it is not yet possible.
These two tables are then used in the next AI prompt to generate an itinerary and a day planning for the morning, afternoon & evening. She improved the outcome by adding Coda references to the text since the AI outputs plain text (something you should keep in mind working with data). The one but last request was a check list for a 4 day trip to Singapore based on the whether forecast. The final part is the optimal order to visit places. She used a pack to generate a map allowing her to drag & drop places on Google Maps.
This is not for starters, but good to follow and most people can generate their own travel logic like this. May you like this example, it might be a good idea to recreate this in your own document. It will give you a feeling of what is possible. That you can create tables filled with content using a written instructing (a prompt) is a kind of magic you really appreciate once you have done it. In the next section an example of how to generate sample data. I urge you to copy — paste the prompt to see what it brings you.
My name is Christiaan and blog about Coda. Since the summer of 2023 mainly about how to Coda with AI to support organisations dealing with texts and templates. Why I focus on Coda AI you can read here: ⤵️
My blogs are for beginners and experienced users. The central theme is that in Coda everything is a list.
I hope you enjoyed this article. If you have questions feel free to reach out. Though this article is for free, my work (including advice) won’t be, but there is always room for a chat to see what can be done. You find my (for free) contributions to the Coda Community and on Twitter.
Coda comes with a set of building blocks ー like pages for infinite depth, tables that talk to each other, and buttons that take action inside or outside your doc ーso anyone can make a doc as powerful as an app (source).
Not to forget: the Coda Community provides great insights for free once you add a sample doc.
