Momento announces $15M Series A funding. Topics event bus released for general availability. Read more →
November 2, 2022 - 6 Min Read

We built a serverless cache you can add to your stack before lunch

With the public launch, our founders reflect on everything that’s gone into Momento.

Daniela Miao
Khawaja Shams


Introducing the Momento team

With public launch, we are reflecting on what got us here. Our mission, as founders, has been to assemble an incredible team and leave a positive impact on the world. We achieve it by improving developer productivity and interactivity of the applications they build. Productive developers lead to rapid experimentation and an acceleration of the global pace of innovation. Time is the most precious commodity and we cherish every second that we can save the world. Instant gratification is no longer a feature, it’s table stakes. For consumers, it means interactive apps that are responsive regardless of the load. The interactivity drives engagement, sustained usage, and conversion. For developers, it means less slogging through operational details and faster time to market for an interactive, production-ready application.

We obsess over the power of the single API call, and what serverless really means for our customers. We obsess over what fast means—beyond just the tail latencies—but holistically and inclusive of the time developers spend to get their applications to production. We obsess over our customers, their feedback, and the fact that they are trusting us in the critical path of their stack. We obsess over the operational rigor that we must uphold ourselves to. We obsess over our culture of free discourse, open communication, transparency, and psychological safety. We argue passionately, and we proudly disconfirm our beliefs.

A little about us (the founders)

We (Daniela and Khawaja) have been working together since 2014 with a shared passion to unlock developer productivity. We have spent our collective careers building mission critical services, making them more observable, and delighting customers by simplifying their stack. DynamoDB routinely serves over 100 million transactions per second, LightStep powers observability and monitoring for the world’s most reliable systems, and NASA colloquially epitomizes mission critical. 

Leading DynamoDB inspired us to deliver large scale systems to customers with a single API call, obviating the nitty gritty details that typically accustom the leaky abstractions in a non-serverless solution. Those details manifest in the form of complex, error-prone configurations that cause distractions with outages, provisioning, and capacity management. 

The problem

The serverless revolution is making time preservation easier—transforming developer productivity by eliminating distractions like provisioned capacity, maintenance windows, inelasticity, and exorbitant price tags. With serverless, developers can finally become less concerned with complex configurations that could result in outages or security risks —and more concerned with good code and great user experiences. Proof of the model’s benefits is in the demand: developers increasingly expect a serverless operational model as the default.

Today, we have serverless compute, storage, databases, queues, and streams. Everything—except caching. That’s a problem. A cache is essential when building interactive applications capable of handling dynamic bursts of traffic. With existing caching solutions, provisioning a cache requires tackling a huge number of error-prone configurations and learning common lessons—one avoidable outage at a time. These misconfigurations come in all shapes and sizes: from underscaling a cache for peak traffic, to not enabling encryption, to using the wrong instance type—but they all result in the same thing: at best, a poor user experience, and at worst, irreparable damage to customer trust.

The opportunity

Caching is in the top five line items in the bill of almost every cloud customer we have run across. Customers are spending billions of dollars each year on caching infrastructure across cloud providers. Building a multi-node caching cluster is time consuming, error prone, and operationally risky. Using dollars spent as the proxy for number of VMs powering caching, and the VMs are a proxy for number of clusters, we see a massive opportunity to save eons of time for customers (and hundreds of millions of dollars) by replacing complex cluster management with a single API call to a cache that works at any scale, without any of the distractions. This is precisely why we did not just build a caching-cluster-as-a-service. We own the end-to-end customer experience for caching, inclusive of the client libraries, a purpose-built protocol, and a serverless backend that handles all the complexity of distributed systems. 

Building the solution

We saw this gap when we observed similar outages across teams at Amazon, AWS customers, and our own experience adding caches to our stacks. Creating a DynamoDB table is a single API call, but adding a cache is measured in sprints worth of work. All that time—distracting developers from experimenting with and innovating on their core business multiplied by the number of caches—adds up. Nothing on the market gets close to solving this problem—and we routinely see incremental improvements in legacy caching solutions in the form of managed services. 

Momento is the first truly serverless cache. It requires *one* parameter: the name of a cache. With a single API call, customers can provision a secure cache that can scale to millions of requests per second—with all the high availability practices built in. Momento sports a radical, uni-dimensional pay-per-use pricing model—one that doesn’t require the mathematical gymnastics to assess what it costs. It circumvents the standard perils of provisioned capacity (pay too much if you overprovision, experience an outage if you underprovision, or both) while committing to the central promise of a robust caching solution: fast out of the box with low tail latencies and high availability, plus security best practices without any configurations.

The proof is in the pudding

Momento is ready for production. Customers like Wyze Labs, Paramount, NTT DOCOMO, and many others are seeing our team’s pedigree prove its worth in our highly available, instantly elastic, and high performing cache to support significant workloads. It’s fulfilling not only to see these customers alleviate their caching woes with Momento, but to simultaneously see them share our passion for improving end user experience.

Our engineers have vast experience building large-scale, mission-critical, performance-sensitive systems at places like Google, Amazon, Valve, Intuit, GitHub—and we cling on to the operational rigor that has enabled us to delight customers with their mission critical workflows at larger firms. Momento customers love our obsession, responsiveness, and relentless passion for operational excellence.

Watching a developer try Momento for the first time is nothing short of inspirational. The bar we set that you can fully onboard Momento faster than it takes to provision a traditional cache cluster. We routinely measure this. Try it. Let us know if it takes you longer than a few minutes to get in and issue your first set.


Momento is here. We have an incredible team, a service that customers are already falling in love with, and a vision to make the world more productive. Our conviction in our mission deepens each time we push a new customer to production. Our multi-tenanted architecture has fundamentally different economics, better availability, and instant elasticity that is unrivaled in the legacy solutions. Our people-obsession—and the obsession of our people—is a virtuous circle between our teammates and the customers. Best of all, we are just getting started!