As you know, just a couple months ago we announced a strategic partnership between Coda and Snowflake, the world’s leading AI Data Cloud company. Our focus is to turn data into action across the enterprise.
At that time, we hinted at something to come—something big.
Well, today we’re excited to share that next big thing, our next brand-new product from Coda: Coda Brain.
It’s a turnkey AI platform, powered by Snowflake, that helps you spend less time hunting for information and more time putting it to work. I like to call it “your team’s favorite know-it-all.”
Coda Brain is the product of a key insight: Consumer AI isn’t the same as enterprise AI. The questions you ask are different. The data returned is different (I’ll come back to this in a second). And most importantly, the information you should have access to is different. Consumer AI is based on the open web; enterprise AI is not. Here’s why:
Coda Brain understands your company’s data. We’re leveraging the ecosystem of 500+ Packs so you can sync all of your relevant data into Coda Brain. Calling all Pack Makers—there are many more Packs to be built!
Coda Brain is secure inside and out. For those interested in AI, we built a RAG-based system that allows for personal authentication for each integration. That means that Coda Brain understands—and respects—data permissions at the most granular level.
Coda Brain is fluent in words and data. We’ve found that many questions at work deserve an answer backed by data. Coda Brain will not only return unstructured responses, but tables, charts, graphs, and timelines as well.
Coda Brain combats hallucinations with verifiable answers. At work, you can’t tolerate an AI system that makes up answers. Coda Brain will produce citations for answers it gives, so you can check out the data for yourself.
Together, Coda Brain and Coda docs become brain-powered docs. We’ve made it incredibly simple to add any result given to you by Coda Brain into a Coda doc—so you can turn one-time queries into continuously refreshing results.
Coda Brain has an accessible API. Your developer and data teams can use API and SQL interfaces to directly connect to Coda Brain: giving them a pre-filled, pre-configured, and pre-encoded data warehouse. In that way, Coda Brain is a platform truly available to your team, not just another application in your stack.
There’s much to love about what Coda Brain will do, so I encourage you to read the full blog post.
Starting today, we’re giving private previews of Coda Brain. Sign up here if you’re interested. And as always, thank you to this community for helping us continue to improve the way people work with Coda.
I’m interested in using Coda to analyze data from my Gmail, Google Drive, and Evernote accounts. Does the Coda and Snowflake partnership announcement indicate any upcoming features or integrations that would make this possible?
Hello. I’m still trying to understand how Snowflake works and the way it charges users. My understanding is that it stores the actual data securely and in Coda we can dictate who has access to the data to add, modify, delete, view, etc. via requests.
However, what is confusing is the pricing…the standard subscription (?) charges $2 per credit but I have no idea what that entails. Is it similar to how AI credits work? Can you please clarify? Thanks.
Hey Alejandro, Coda Brain is the tool you’ll use to do that! You connect your Evernote, Drive, and Gmail accounts, and then you can query that data with Brain.
Hi @Coda_Solutions, thanks for the great question. Snowflake has a unique “consumption based” billing model. In short, they charge you for storage (holding your data) and compute (working with your data). This is similar AI credits in Coda where you pay based on your usage.
I recommend reading Snowflake’s pricing guide for a detailed explanation of how exactly this works, but the tl;dr is that you pay based on how much you use their services.
Hi @Piet_Strydom! When launched in GA, we intend for Coda Brain to be available on all tiers. Just because it’s built with enterprise-grade security doesn’t mean smaller businesses shouldn’t get access .
Hey @Chris_Williams! The Private Preview will generally be limited to Enterprise customers, but please fill out the interest form as we will review interest on a case-by-case basis.
I don’t have a timeline I can share for GA at the moment, but we’ll keep the community up to date as things develop.
Yes and no. That is because it looks like the Snowflake engine sits on top of the packs so whatever info you pull in coda through packs then you can query it with Coda Brain and have it.
Non-IT data workers can now use the wonderful no-code capabilities of Coda and the impressive AI strengths of Snowflake to ask deep questions of their corporate data and have those questions answered, as rich text, as tables, and as charts! And then use those answers in their collaborative, Coda documents, to turn data into actions, and track progress across their teams.
This opens a new category of AI enabled technology to the information workforce, one that promises significant economic returns on investments in data and knowledge. This wont take away our jobs, it WILL make our work way more effective and productive.
Hey @AJM! Currently, the Snowflake Pack (our integration that allows you to query
Snowflake data directly in your docs) is only available to the enterprise tier. Coda Brain, however, will be available for all customers when it goes into GA.
If you’re interested in the Pack but not on enterprise, send me a private message—we can explore options!
Hey @Cristian_Nichifor! Those are some great questions with nuanced answers. On the first, the tl;dr is that tables returned in Coda Brain are regular Coda tables you can group, sort, format, and so on. On the second, we intend to keep the data as up to date as possible, but given it needs to be ingested and embedded for AI, it will not be quite real time. Feel free to reach out privately if you’d like to discuss in more detail.
How does this work with coda’s current row/size performance limit? I feel limited already with how big my tables get how will this work where you want more data into Coda?
.