Sphere

The best people in tax tech are voting with their feet. Ben Lewis has spent 15 years in tax technology, most recently as Director of Tax Platform at a competitor. As of today, he's building with us. Same reason Alexander Kendall left PwC in August: he looked under the hood of TRAM and saw what tax looks like when AI is the foundation rather than a bolted-on afterthought. In tax tech, ‘legacy’ now means anything built before ChatGPT. Ben knows the platform side of global tax better than almost anyone, having worked with and advised some of the biggest companies in the world. Exactly the person to help drive Sphere in EMEA. The guard has changed, and the jerseys are better too!

Tax teams don't need one more system to babysit. They need a partner who shows up when the deadline is real and the stakes are higher than just another support ticket. This G2 Fall 2026 season, our customers made that case for us. Four badges: Momentum Leader, Easiest Setup, Best Support, and Easiest To Use. Thank you to every customer who took the time to share their experience.

While interviewing at Sphere for TRAM Lead role, I asked the question: "What's the gap between how the product is marketed and what it actually does?" The answer I got was that there wasn't any. Sure. Every product team says that. So I spent my entire first week doing what any reasonable new joiner would do: quietly interrogating every corner of the product to find the gap. But I found... nothing. The product is, annoyingly, exactly what we say it is – internally and externally. Which, if you've worked in tech marketing for more than five minutes, you know it doesn't happen very often. TRAM is the product I lead - and no, it's not an actual tram. No overhead cables, no metal tracks, no city streets involved. Just tax tech. Sorry to disappoint the public transit enthusiasts. What does TRAM do exactly? It monitors tax changes and does the research for you, pulling from tens of thousands of legal sources across the globe - all verified, not vibes-based. We still keep a human in the loop before anything touches the tax engine, because "fully autonomous AI tax decisions" is a sentence that should scare everyone. This is also why you won't catch us sprinkling "AI" onto every button and page title like glitter. We built the real thing from day one. It doesn't need a marketing touch-up. TRAM can actually conduct research across hundreds of product sub-categories, capturing virtually every tax-relevant variation globally. Our recent record is 348 distinct sub-categories for a single product. Manually, that's at least a multi-month project + a constant maintenance. With TRAM it took us one week – from the concept to go-live. We're not trying to brag. ...okay, we're a little bit bragging. The future's here. Stay tuned for what's next. Or ping me to hear more!

There’s a common question that comes up in almost every demo call we have at Sphere: “How do you make sure your AI is getting our tax right?” The simple answer is: our AI research model isn't the same as your regular ChatGPT/Claude search. Tools like ChatGPT and Claude weren't built to reason over constantly changing tax law - they’re typically trained on a snapshot of the internet. Tax law is a living and breathing beast. It’s jurisdiction-specific and full of edge cases that don't make it into training data. So when you ask Gemini a tax question, you often get a confident answer that was maybe true, somewhere, at one point in time. Here’s how TRAM (our AI-native tax research system) is different: → It continuously ingests primary source legislation, guidance, bulletins, and rulings directly from tax authorities - not the internet generally (so it’s not learning from what some random guy said on Reddit 4 years ago) → It uses fine-tuned reasoning models trained specifically for multi-step legal and tax reasoning → Every determination it makes comes with citations back to the actual source material → Before anything goes live into our tax engine, a Sphere tax expert reviews and approves it We’re not replacing humans, we're just making tax research faster, more reliable and scalable. TRAM does the heavy lifting of finding, reading, and reasoning over the law, then our experts spend minutes, not days/weeks reviewing each determination. I use Claude every day - for work research, planning my SF weekend getaways, finding recipes and more. General LLMs are great at many things but high-stakes tax research isn't one of them. [TRAM info in comments]

Not all integrations are created equal. And that's why we're incredibly excited to announce Sphere's NATIVE integration with Stripe. With this new integration, Sphere will have native API access into Stripe Billing and Checkout products which provides the following benefits: 🧱 No-code, instant setup (literally flick a toggle in your settings vs setting up webhooks) ✅ No errors and admin headaches (we have direct API access vs other solutions which use webhooks that break all the time) ⚡️ Immediate tax calculation on invoices AND checkout (lightning fast speed, pushed straight into Stripe tax objects) 💵 Accurately accommodate tax implications of refunds / credit notes (which other vendors cannot do) This was built in conjunction with the Stripe Billing and Checkout team to be a completely seamless, plug-and-play experience. And it was announced today at their Sessions conference! Want to get started? Download the Sphere app in the Stripe App Marketplace today: https://lnkd.in/gmKa_vZ2

How did Replit scale their finance function while going from $2M to $250M in just 2 years? The narrative of companies hitting $100m ARR in the space of 12 - 24 months is becoming more common than not. Lovable, Ramp, ElevenLabs, Replit. The most crucial, yet often overlooked common denominator in their success is the focus on scaling back-office infrastructure with the same urgency as product. But what systems and processes enable that sort of growth? We'll be hosting a chat with Tim Ryan (Senior Accounting Manager at Replit), Kaitlyn Sohn (Foroud) (Sphere's Head of Tax) and myself on 27 May. We'll get into: • The build vs buy decision • How to retain a lean finance team as the company scales to $9B+ valuations • What's changing in international tax compliance for SaaS companies • How to think about data structure and procurement as you scale One more thing those companies have in common, they’re all Sphere clients. Let me know if you’re interested in us addressing anything in particular and we can work it into the session. Link: ordnl.link/g5KsOkx