Fireblocks Unifies Data and AI to Empower Users and Create Stickier Customers

While data growth was front of mind for Fireblocks, it also needed to ensure its data ecosystem could be unified to deliver comprehensive insights to customers and internal users. However, with 15 different data domains, each with its own data and semantic layer, this wasn’t a given.

“Snowflake has been really valuable here as it gives us a way to cross domains,” says Swissa. “With everything flowing into Snowflake, we can serve anyone in the business — whether it's a sales rep that needs revenue insights, or a product developer looking for data on staking accounts.”

Crucially, customers can now access this asset and transaction data securely thanks to Snowflake features like RBAC and row-level security. “At first, I didn’t think we’d be able to do this without a massive risk of data leaks,” says Swissa. “But then in a 45-day hackathon, we developed our AI agent, Fire Genie, on Snowflake. That lets customers query only their data in a secure environment through an API to get real context on data across all their wallets and assets.” This is vital in helping customers maintain visibility of the performance and security of their cryptocurrencies, stablecoins and other digital assets.

To ensure accuracy, Fireblocks designed Fire Genie to clarify context and ambiguous terms with users. For example, MATIC, a proof-of-stake scaling solution, upgrading and changing its name to POL has led to many users using one term or the other for the same thing. In these cases, Fire Genie will clarify to ensure it’s surfacing the right insights to users and delivering the most accurate data at all times.

So far, the preview tests for Fire Genie have exceeded expectations for user engagement, delivering accurate responses that customers then use to ask follow-up questions. “The customer experience is crucial for us,” explains Swissa. “By offering a more personalized set of analytics to customers through Fire Genie, these clients are likely to stay with us longer.”

After seeing the positive responses to Fire Genie, the team decided to bring the power of AI to internal users. Fireblock’s internal agent does exactly that. Powered by Cortex Agents with semantic views, it offers all users a way to easily query data in natural language.

For example, product teams can now easily interrogate what parts of the product are landing well with customers, and what features might be causing frustration. “With our internal agent and Snowflake, teams can go from a Jira ticket to the deepest analytics on product metrics in one place — all with correct semantic views,” says Swissa.

By using Streamlit together with Snowflake on AWS, internal users can also quickly build custom reports and share them with other departments. “In the past, teams would be sharing static screenshots that would quickly become out of date,” recalls Swissa. “With Streamlit and Snowflake, our teams can now share live, up-to-date feeds so everyone can make better decisions based on a single source of truth.”

So far, this automated approach to analytics has democratized access for 200+ users, while saving significant time across the data team. Fireblock’s internal agent now powers over 2,000 analysis queries per month, roughly 40-50% of all queries. This saves the equivalent of two full time analysts’ worth of capacity every month. 

“Our analysts used to struggle to meet demand for queries,” says Swissa. “Now, they’re freed from basic tasks and can focus on delivering the kinds of advanced insights that lead to true product improvement, customer retention and higher revenue.”

Fireblocks’ internal and external AI agents are only just getting started. For the external Fire Genie, Swissa and his team are looking to expand functionality to allow clients to create their own personalized dashboards so they can dig deeper into their data — further improving the user experience and making customers even more loyal to Fireblocks.

And for its internal agent efforts, the digital asset infrastructure provider wants to create a one-stop-shop for all in-house analytics queries. “With everything available in Snowflake, we want any user to be able to self-serve any analytical query — from initial idea to complete final report,” explains Swissa.

“This is the kind of feature that helps us make impactful decisions that allow people to make more of their digital assets, and do even more with our platform. And that keeps clients working with us, and helping us grow sustainably.”