Malachyteprovides behavior intelligence infrastructure for e-commerce, creating continuously updated shopper profiles from real-time activity to power search, recommendations, and product pages. Pulse 2.0 interviewed Malachyte Co-Founder and CEO Sidd Motwani to learn more.
When asked about his background and the experiences that shaped Malachyte, Motwani shared:
I’ve effectively spent my whole career on one problem: how systems come to understand people based on their behavior.
I built user data systems from zero at Priceline, then led product for Spotify’s user representation platform from 2019 to 2023, the layer underneath Discover Weekly, Home, Search, and Radio for more than 600 million people.
A lot of that was building tools for merchandisers and executives rather than only for engineers, which shaped how Malachyte is designed today.
The pull toward technology is probably inherited. I grew up around entrepreneurs, and my late uncle, Dr. Rajeev Motwani, co-created PageRank, so early on I watched how one invention created the infrastructure that changed how the entire world found information.
That leaves an impression: the layer underneath usually matters more than the interface sitting on top of it. Malachyte is our way of building that layer for commerce.
When discussing how the idea for Malachyte came together, Motwani explained:
Ian and I spent years at Spotify focused on one problem: proactively predicting what a user actually wants to listen to right now based on their intent and likely next action, rather than purely on their listening history.
With 800 million users and a billion-song catalog, solving that meant building a different kind of recommendation technology based on two-headed user vectors.
The two-headed model indicates two things about a listener at once:
- Long-term taste, or the slow head: what they generally gravitate toward over weeks and months, tuned toward keeping them subscribed and discovering new music for years, which is the metric that matters for lifetime value.
- In-the-moment intent, or the fast head: what they’re actually doing right now through their behavior on the app, which can shift by the minute.
The hard part was reading both simultaneously and being able to adapt fast enough to matter.
Solving this led to an interesting insight. In the same way the two-headed vector reads intent for known listeners, it also accurately predicts intent for a brand-new listener with no history. In e-commerce, this is known as the “cold start” problem.
Armed with this new approach to cold start, Ian and I started talking about our own personal digital experiences and where they felt frustrating or broken. Retail stood out immediately.
It’s the one large category where most of your traffic is a stranger, and nobody had built anything to handle cold start in real time. Instead, what retailers had was batch technology: a catalog and a set of rules refreshed overnight, not a system reacting to what’s happening on the site right now.
The idea for Malachyte was born – which was to apply a two-headed user vector in a novel way.
When asked about his favorite memory from building Malachyte, Motwani recalled:
The first time Malachyte ran in production on a live store was in the fall of 2025.
There’s a particular feeling you get when you witness a model you’ve argued about internally for years start learning from real shoppers you’ll never meet. It either works in the wild or it doesn’t, and there’s nowhere to hide.
And it worked!
Another moment that stuck came a little later, when we ran head-to-head against a customer’s existing stack.
We found that the largest lift in revenue per visitor showed up among brand-new visitors, people the system had never seen before. That was the claim we’d been making from the beginning and the one everyone was most skeptical of.
Watching it hold up on someone else’s traffic, measured by someone else’s team, was worth more than any demo we ever gave.
When describing Malachyte’s core products and features, Motwani detailed:
Malachyte offers Behavior Intelligence Infrastructure for e-commerce.
We build a live profile of every shopper, including the roughly 90% who never log in, from what they do in the moment: what they search for, click on, compare, and skip past.
That profile updates continuously and powers search, recommendations, and product pages, so the experience adapts while someone is still on the site rather than after they come back.
For brands, it shows up as revenue per visitor. Their merchandising team also gets no-code control over the live model, so they can promote inventory or change what it optimizes for and see the effect in minutes instead of after tonight’s batch job.
We install it for you in seven days, and it deploys on Shopify after a 21-day pilot with no replatforming.
When asked about recent challenges in the e-commerce personalization sector and how Malachyte has addressed them, Motwani noted:
There are two, and they’re connected.
The first is that “AI personalization” has been promised to retailers for the better part of a decade and has mostly underdelivered, so there’s real fatigue in the market.
We deal with that by not asking anyone to take our word for it. We run head-to-head against whatever technology the brand already uses on its own traffic with significance testing and let the result settle it.
The second is trust, which affects all of the interconnected relationships necessary to see results, and it’s warranted because somany solutions require cookies or login and AI recommendations lack explainability as if they come out of a black box.
Shoppers are tired of pop-ups and consent banners, and brands are careful about where their customer data ends up.
Our answer is structural rather than reassuring: We read how someone behaves, not who they are. No login is required, there are no third-party cookies, and there is no hoarding of identity.
Each brand’s model is trained on its own traffic rather than pooled into anyone else’s.
To avoid the perception that our recommendations are delivered from a black box, we show merchandisers how the user vector is driving ranking. We allow them to adjust Malachyte outputs if there are merchandising strategies that take priority over pure behavior-driven rankings.
Trust turned out to be a product problem, not a messaging one, and we’ve addressed it for both consumers and business users.
When discussing how Malachyte’s technology has evolved since launching, Motwani described:
The model itself has changed very little. We knew the two-headed architecture was the right approach early. What’s changed is who can actually use it.
In the beginning, this was a powerful system that needed us in the room to make it work. Now, we’ve productized it for scale.
We install it for you in seven days, and it deploys on Shopify after a 21-day pilot with no replatforming. One shopper profile drives search, recommendations, and product pages instead of three separate models pulling in different directions.
The biggest addition came straight from customers: our Merchandising Studio, which gives business teams no-code control over a live model.
They can promote inventory or change what the system optimizes for and see the effect in minutes rather than after tonight’s batch job.
Customers kept telling us that a system they can’t see into or steer isn’t a system they’ll trust. They were right, and building for that changed the product.
When asked about Malachyte’s most significant milestones, Motwani said:
I mentioned two special milestone moments above. In addition to those, the $10 million raise is also very significant because it means we can continue building products, serving customers, and scaling in a meaningful way.
When invited to share specific customer success stories, Motwani explained:
So far, our results with early customers have been incredibly exciting.
We’ve produced a 31% lift in revenue per visitor at Fun.com, an 80% lift in add-to-cart click-through at Brunt Workwear, and a 17% lift in revenue per visit from new visitors at Jordan Craig, while holding latency below 200 milliseconds through Cyber Monday traffic.
When asked about Malachyte’s funding, Motwani stated:
We raised $10 million in seed funding, co-led by Bessemer Venture Partners and Gradient Ventures, with participation from Harpoon Ventures.
When discussing the total addressable market Malachyte is pursuing, Motwani outlined:
We sell to consumer e-commerce brands and retailers, from mid-market direct-to-consumer companies through large, established retailers.
The immediate market is what brands already spend on search, recommendations, and personalization software, which is a multibillion-dollar category. But that understates it because those are line items on a website budget.
What we’ve built sits underneath merchandising and, over time, marketing spending as well, which is a far larger pool.
The shift toward AI agents handling routine buying makes real-time behavioral understanding infrastructure rather than a feature. McKinsey has put agent-orchestrated retail spending at $3 trillion to $5 trillion by 2030.
Beyond e-commerce, we see other use cases in travel, e-learning, streaming, and other categories where this behavior intelligence infrastructure will be crucial in the future.
When asked what differentiates Malachyte from its competition, Motwani emphasized:
While we weren’t trying to solve for “cold start” at Spotify, it turns out that our novel application of user vector technology not only reads long-term taste but also predicts in-the-moment intent.
This is exactly what is needed for e-commerce use cases: preference and intent prediction based on real-time behavior.
We call it behavior intelligence, and it beats demographic and other forms of profiling for anonymous users every single time.
The challenge is that commerce signals are not captured. Most e-commerce platforms are built to track identity logins, cookies, and purchase history, not real-time behavior.
The stakes are also higher for e-commerce than for music. In this industry, a wrong recommendation costs a sale, not just a skipped track.
Inventory and pricing shift daily, and shopping has a purchase goal, so optimizing for a different level of engagement is essential. In e-commerce, visits that are weeks or months apart instead of back-to-back days or hours apart require a robust system that learns.
Unlike most AI solutions, we are not using a large language model. We’re not predicting the next word in a sentence. We’re using vector AI to predict the next product a shopper wants from real-time behavior.
We leverage our own proprietary two-headed transformer. One head is a slower base model that learns a shopper’s general taste over time. The other fine-tunes continuously based on what they’re doing in this exact session.
Most personalization systems have a version of the first but nothing like the second, which is why they require a login to know who you are but lack insight into what you’re doing right now.
This is the hard part: training a model that learns in real time and serving it to every shopper fast enough that nobody notices.
Point-solution vendors, such as Bloomreach and Algolia, are solving real, adjacent problems like search and merchandising.
We don’t think we’re in the same category as them, and a feature-by-feature comparison undersells what we’re actually building.
Most require user logins or cookies to personalize results, and frankly, consumers are tired of the barrage of pop-ups, spending half their shopping experience closing them or wondering what personal information is being tracked without their knowledge.
We built Malachyte so any brand can access cutting-edge technology trained on its user traffic without having to invest tens of millions of dollars in computing and building deep expertise in behavior intelligence.
When discussing Malachyte’s future goals, Motwani concluded:
Longer term, the same real-time understanding of a shopper that decides what they see should also inform what a brand spends to reach them.
Those run as separate systems today, on much coarser signals, and they shouldn’t.
Bringing merchandising and marketing onto one behavioral profile is where this goes.
While we focus on retail today, we’ll also expand across digital categories, including travel, finance, gaming, and more.
