Account Executive

Canary Data and Perplexity have partnered to integrate Canary's investment research platform into Perplexity's answer engine and Perplexity Computer. The partnership allows joint clients to access Canary's proprietary datasets, contextual layer, and AI analysts directly within Perplexity's environment. Canary combines dozens of proprietary datasets — including undisclosed business events, AI disruption exposure, and investor idea pitches — with alternative data such as credit card transactions and web traffic. AI adds context to make the data useful for investors. Perplexity users can now query Canary's datasets, work with contextualised versions, or request finished analysis from Canary's AI analysts. The companies are launching with a bring-your-own-licence model, enabling clients to connect existing Canary entitlements to Perplexity without additional API configuration. Canary was founded by a former Tiger Global partner and is backed by Tiger Global and Arena Holdings.

Firebird launches Nvidia-powered 300MW AI data centre in Armenia. - John Tanner - 11 August 2026 - 2940 views Nvidia announced on Saturday that AI cloud startup Firebird has launched a new AI data centre in Armenia, powered by Nvidia GPUs and Dell Technologies' AI infrastructure. Firebird's AI data centre in Hrazdan - which was delivered in six months - was built on Nvidia's AI factory platform, which includes reference architecture, GPUs, networking and AI software. The facility hosts over 70,000 Nvidia Rubin and Blackwell GPUs on Dell PowerEdge servers, with its AI infrastructure supporting an IT load capacity of 300MW. The data centre also runs on Nvidia's DSX platform, which integrates computing, networking, power and cooling as one codesigned system. Firebird said that enables it to run up to 40% more GPUs on the same footprint, producing more tokens per dollar and extracting more value from every megawatt of capacity Schneider Electric provided the power infrastructure supporting the AI data centre, including medium- and low-voltage switchgear, three-phase uninterruptible power supply systems and rack enclosures to make sure the power buildout keeps up with pace of compute. Meanwhile, Vertiv has provided a cooling architecture combining chilled-water technology, advanced controls and Vertiv TrimCooler technology for efficient heat rejection. The cooling resources are centrally coordinated by Vertiv's chilled water manager to improve visibility, efficiency and responsiveness as demand shifts with AI workloads. Nvidia said Perplexity AI has already signed on as a customer, using Firebird's high-performance AI infrastructure for its AI agent platform and answer engine. Nvidia said the Hrazdan AI data centre is the largest in the CIS, and will help make the region a magnet for global companies building and running AI, as well as a catalyst for local developers, researchers and enterprises. Firebird co-founder Alexander Yesayan said the Armenian data centre is part of its broader ambition to pursue an AI infrastructure roadmap spanning Armenia, Kazakhstan and additional markets. "Our ambition for what we are building in the next two years or so is roughly 2 GW of capacity around the world," Yesayan said. "We're very focused on merging into frontier markets." Firebird also announced that Nvidia intends to invest in the company, following an earlier investment by CoreWeave this year, both of which will help finance that expansion plan. No details were given on the size of the Nvidia investment.

AI shopping search added virtual try-on. What it means for your store. Quick answer Google and Perplexity have both added native virtual try-on to their AI shopping search results, letting a shopper upload one photo and preview clothing before buying, right inside the search or chat interface. That shifts what shoppers expect, but both only work on products those platforms have already indexed inside their own apps. For a fashion store, the practical takeaway is that shoppers now expect to see themselves in a garment before buying, and the more resilient way to meet that on your own product pages, ads, and in person is to run virtual try-on yourself rather than depend entirely on a platform feature outside your control. Virtual try-on used to be something a shopper found on a handful of fashion sites. In the past year it has shown up somewhere much bigger: inside the AI shopping tools people already use every day. Google and Perplexity have both shipped their own try-on features directly into search and chat. That is new, and it changes what shoppers expect, even from stores that never install anything from either company. What Google and Perplexity actually launched. * Google's Try-On tool lets a shopper upload a single full-length photo and generates images of themselves wearing apparel across Google's Shopping Graph. Results can appear in Search, Google Shopping, and Google Images, and the feature has expanded to a growing list of countries, including the US and the UK. * Perplexity's Virtual Try On rolled out to all Perplexity Pro and Max subscribers. A shopper uploads a photo once to build a full-body avatar, then a try-on button appears next to supported products while shopping inside Perplexity. Perplexity has said the feature runs on Google's Nano Banana image model family, the same family (Gemini 3 Pro Image, also known as Nano Banana Pro) that powers Corlen's own try-on engine. It currently requires a Pro or Max subscription and is limited to the US. * ChatGPT, by contrast, has not shipped an equivalent in-chat try-on. OpenAI's Instant Checkout, a different feature aimed at completing a purchase inside ChatGPT, was pulled back in March 2026 after fewer than 30 of Shopify's millions of merchants ever went live with it. ChatGPT's shopping experience today is discovery first: it surfaces products, then sends the shopper to your own store to actually buy. How the built-in try-on features actually work. Both Google and Perplexity work the same basic way a shopper never has to think about: upload one ordinary photo, no 3D scan or in-store visit, and an AI image model renders the chosen garment onto that photo. The result is generated fresh for each item a shopper taps. Where they differ from a store's own try-on is where that result lives: entirely inside Google's or Perplexity's own interface, tied to whichever products that platform has already indexed for the feature. What this means if you run a fashion store. None of this requires a store to do anything to keep functioning. But it changes the baseline shoppers now compare every store against, and a platform feature does not give you everything: * The expectation bar just moved. A shopper who has tried Google's or Perplexity's try-on feature once now knows what a good preview looks like, and will notice its absence on a product page that only shows a flat-lay photo and a size chart. * Being indexed is not fully in your control. Whether your catalog even shows up for Google's Try-On tool or Perplexity's try-on button depends on your product feed being picked up correctly by that platform's own shopping graph, not on a decision you make directly. * It only runs on that platform's surfaces. A shopper who tries on your product inside Google or Perplexity still has to leave that interface, land on your actual store, and buy there. The try-on itself never touches your product page, your ads, or a customer standing in your physical store. * Platform commerce features can change fast. OpenAI's Instant Checkout shipped and was pulled back within about six months. A feature you have no control over is not something to build your customer experience around exclusively. Should you rely on platform try-on, or add your own. These are not really competing options. A shopper discovering your product through Google or Perplexity benefiting from their built-in try-on is a good thing, not a threat. The gap is everywhere else: your own product pages, your ads, your email campaigns, and, if you sell in person, the shop floor itself. None of that is covered by a feature that only exists inside someone else's app. This is the exact gap Corlen is built to close. It adds virtual try-on directly to your own Shopify product pages, so a shopper does not need Google or Perplexity to see themselves in an item. The same engine also runs a try-on kiosk for a physical store, and a developer API for a team building a custom flow, so the experience is not limited to whichever AI shopping assistant happened to index your catalog that week. For the underlying business case, see how to reduce online fashion returns, since a shopper who sees an accurate preview before buying is also a shopper less likely to send the item back. Getting started. You do not need to wait on any platform's roadmap to give shoppers this experience today. If you are comparing options, how to choose a virtual try-on app for your store covers what actually matters beyond a feature list. Or see it directly: try Corlen on your own photo in a few seconds, no account needed, or check the Shopify integration if you want it on your own store. Frequently asked questions. Does Google's AI shopping search have virtual try-on? Yes. Google's Try-On tool lets a shopper upload one full-length photo and see themselves wearing apparel listings across its Shopping Graph, in Search, Google Shopping, and Google Images results. It has expanded to a growing list of countries, including the US and the UK. Does Perplexity have a virtual try-on feature? Yes. Perplexity rolled out Virtual Try On to all Pro and Max subscribers, letting a shopper upload a photo to generate a full-body avatar, then tap a try-on button next to supported products while shopping inside Perplexity. It currently requires a paid Pro or Max subscription and is limited to the US. Is ChatGPT's shopping feature the same as virtual try-on? No. ChatGPT's shopping tools currently focus on product discovery rather than an in-chat try-on. OpenAI also pulled back its Instant Checkout feature in March 2026 after fewer than 30 of Shopify's millions of merchants ever went live with it, a reminder that platform-level shopping features can change or disappear quickly. Can I add virtual try-on to my store without waiting for a platform to support it? Yes. Corlen adds virtual try-on directly to your own Shopify product pages, an in-store kiosk, or a custom flow through the API, so the experience runs on your storefront regardless of whether a particular AI shopping assistant has indexed your catalog for its own try-on feature that week. Do Google and Perplexity's try-on features work for every online store? No. Both only work on products already indexed inside that platform's own shopping feed or Shopping Graph, and both currently have geographic or subscription limits. A store cannot rely on either one to guarantee coverage of its full catalog for every shopper. Ready to add real try-on to your store? Install Corlen on Shopify in minutes, or build it into your own platform with the developer API. See it on your own photo Corlen shows your customers how any garment looks on their own body, in seconds. Try it free, no account needed.