Sendra has launched on Product Hunt with a pitch that sounds like a time capsule from 2016: design emails in Figma, export HTML that works everywhere. The description is two clauses long and the promise is the oldest in the email business. Yet the timing is anything but retrograde. Sendra is arriving at the exact moment when the email stack, long the most stubbornly hand-coded corner of the design world, is finally being pulled into the same automated pipeline that swallowed landing pages a decade ago.
The pitch itself is familiar. Figma became the default canvas for product design because it gave designers a single source of truth. Email never got that treatment. Mailchimp, Klaviyo, and HubSpot each ship their own drag-and-drop builders, each with its own quirks, each exporting HTML that breaks in Outlook or clips in Gmail. Designers who want pixel fidelity end up in Figma anyway, then hand the art to a developer who rebuilds it in table-based HTML with inline styles, because that is the only markup the email clients will reliably render. Sendra’s bet is that the rebuild step is the waste.
The interesting question is not whether the export works. It is why this is happening now, and what it says about where the rest of the design toolchain is heading.
Email HTML is a hostile environment. The rendering engines are frozen in time. Outlook uses Word to render HTML, which means it ignores flexbox, ignores grid, ignores CSS variables, and occasionally ignores basic sanity. Gmail strips <style> blocks and forces inline styles. Apple Mail is comparatively modern but still lags the browser baseline. The result is a discipline that survived by memorizing quirks: margin hacks for Outlook, bulletproof background buttons, MSO conditionals wrapped in <!--[if mso]> comments. It is the closest thing the web has to assembly language, and it is exactly the kind of work that generative AI and automated tooling are now good at.
Sendra is not alone in noticing. The email-design tooling space has seen a quiet wave of entrants over the past two years, all chasing the same wedge: take the design out of the code and let the tool generate the hostile markup. What makes Sendra’s Product Hunt listing notable is the explicit Figma-native positioning. It is not a standalone editor pretending to be a design tool. It is a plugin or a workflow that sits inside the tool designers already use, which is a materially different go-to-market strategy. The hard part of email design was never the design. It was the translation layer, and Sendra is claiming that layer as its product.
There is a pattern here that extends well beyond email. The broader design-tool economy is consolidating around the idea that the artifact designers produce should be the final artifact, not a blueprint for someone else to rebuild. Figma’s own Dev Mode pushed in this direction by trying to make the handoff to engineers less lossy. AI code generators like v0 and Lovable took the next step, generating production UI from natural-language prompts. Sendra is applying the same logic to a domain that resisted it for two decades, not because the problem was hard to see, but because the constraints were so ugly that nobody wanted to build the tooling.
The economics are real. Email remains the highest-ROI channel in digital marketing, and the cost structure is dominated by labor. Every campaign that needs a custom template requires a designer, a developer, a QA pass across clients, and a round of fixes when Outlook inevitably breaks something. Automating the export step compresses that pipeline. It does not eliminate the designer, but it eliminates the developer bottleneck and the QA ritual. For a marketing team shipping weekly campaigns, that is a meaningful line item.
The AI angle cuts deeper than automation. The reason email HTML is expensive to produce is that it requires encoding design intent into a constrained, hostile format. That is precisely the kind of translation task that large language models have become unexpectedly good at. Sendra’s export engine, whatever its implementation, is standing on the same underlying capability that powers code generation everywhere else: the ability to map a high-level visual description onto low-level, syntax-heavy output. The email client quirks are a feature, not a bug, for this kind of tooling. They are a fixed, well-documented set of constraints, which is exactly what a model can be trained to satisfy.
None of this is to say the launch is a guaranteed win. The email-tooling graveyard is full of products that solved the technical problem and died on distribution. The buyers are not designers. They are marketing operations managers who already have a stack, and the stack is sticky. Klaviyo and Mailchimp bundle their editors into platforms that also handle deliverability, segmentation, and analytics. A Figma plugin that exports HTML still needs a sending platform, and the sending platforms have little incentive to make their built-in editors obsolete. Sendra’s path to revenue runs through the gap between what the platforms offer and what serious design teams want, and that gap is real but narrow.
The more consequential read is what this signals for the design-tool economy at large. If Figma-native email export works, the same pattern is portable. Newsletters, invoices, receipts, and any other document type that still lives in hand-coded templates are all candidates for the same treatment. The design artifact becomes the source of truth, and the generated code becomes a compile target. That is the direction the entire industry has been moving, and Sendra is one more data point confirming it.
The deeper implication for AI builders is about where the value accrues. The tooling layer is getting commoditized fast. Any competent team can build a Figma-to-HTML exporter now, because the hard part was never the exporter. It was the years of accumulated knowledge about email-client quirks, and that knowledge is now encoded in models and in open-source libraries. The moat is not the translation. The moat is the workflow integration, the QA automation, and the trust that the output will render correctly in every client on the first try.
Sendra’s Product Hunt page does not say how it handles Outlook’s Word rendering engine, or whether it supports MSO conditionals, or how it tests across clients. Those details will determine whether this is a toy or a tool. But the direction is clear. The last hand-coded corner of the web is being automated, and the tool that does it will look less like a code editor and more like a compiler.
The email template was the last place where designers still had to think in tables. Sendra is betting that the next generation of designers will not have to think in tables at all.