Tom Invernizzi
25 posts

Tom Invernizzi
@Tom_inv
Building Combinely (YC X25)

Every tax accountant knows the grind of pulling messy, complex client data and re-entering it into tax software, line by line, under time pressure. K-1s don't look the same twice. Different partnerships, different formats, different footnotes. That variability is exactly why tax teams spend 2-4 hours per client manually entering data into CCH, Lacerte, and UltraTax. The footnotes? That's where the real complexity lives. Dense, critical, error-prone. Solutions are out there. They're part of the solution. But most of them miss the details. We want to change it. Watch our AI coworker handle it: ↳ Upload K-1s, 1099s, W-2s, Schedules all at once ↳ AI extracts every detail, regardless of format ↳ Runs calculations (Schedule C expenses, rental income, deductions) ↳ Data flows into your tax software automatically ↳ 2-4 hours becomes minutes See it working live: combinely.ai



Last month, we had 70+ builders in our SF office for Sonar API Demo Night! We saw 8 incredible demos featuring AI accounting tools, smart real estate assistants, real-time brand monitoring, and more, all built on our Sonar APIs. Didn't get a chance to demo? Join our brand new dev community and share your project and thoughts about our APIs! (link below)

Many AI apps today feel like the "horseless carriages" of the late 19th century, which "swapped a horse for an engine without redesigning the vehicle to handle higher speeds." They pack powerful tech into outdated interfaces. YC's Pete Koomen (@koomen) thinks we can do better. He joined @t_blom and @dflieb to lay out a new vision for how AI should actually work: not as a chatbot bolted onto legacy software, but as a customizable tool that helps people offload the work they don't want to do. From editable system prompts to agents that act more like collaborators, they break down what it means to build AI-native software—and why the future belongs to products that let users teach machines how they think. 0:00 – Intro 0:52 – Why AI apps are broken 2:39 – How Gmail’s AI features fall short 4:00 – A better way to build AI apps 5:27 – The hidden system prompt 7:57 – What if you could access the system prompt? 9:40 – The developer-user divide in software 10:48 – The "horseless carriage" metaphor 13:35 – Email reading agent demo 14:34 – Everyone can be a prompt engineer 16:23 – Why coding agents feel magical 21:42 – Training AI like a human assistant 28:45 – The problem with chatbot interfaces 29:10 – Advice for founders

.@Combinely is building an AI-coworker for accountants. It reclaims the 30-40% of an accountant’s day that disappears inside Outlook – chasing docs, digging for context, and answering client queries. ycombinator.com/launches/NPb-c… Congrats on the launch, @Tom_inv and @ArthurGranacher!





🚀 @Combinely launched! AI Coworker for Accountants "AI-coworker that helps accountants review & produce work." 🌐 fondo.ai/430ZY0l Founded by @Tom_inv & @ArthurGranacher

.@Combinely is building an AI-coworker for accountants. It reclaims the 30-40% of an accountant’s day that disappears inside Outlook – chasing docs, digging for context, and answering client queries. ycombinator.com/launches/NPb-c… Congrats on the launch, @Tom_inv and @ArthurGranacher!









