Launching today
Jupid
File your taxes with Claude Code
155 followers
File your taxes with Claude Code
155 followers
No matter how powerful LLMs get, they are objectively bad at financial transactions. Context loss, inconsistent categories, no memory between sessions. Jupid fixes the data layer. Connect your bank — it learns your business and every vendor relationship once, then remembers forever. Transactions mapped to IRS Schedule C categories (~96% accuracy). Missed deductions found: $1,249/year average. File your Schedule C in 5 minutes. Works with Claude Code. Free trial + 50% off first 3 months.
@slavaakulov How does Jupid "remember" that context long-term without constant retraining, especially for solopreneurs like me juggling multiple clients?
Sidekick Browser for Teams
@slavaakulov
This is fantastic, after years of IRS fines, missed deadlines and accounting mess (there's always something more important to do in a startup than accounting) having something like is a godsend :) I only wish you guys launched a decade ago )
Spiritme
What about using an LLM with a vector database for persistent context? Seems like that would solve the memory problem without a new product.
@nikita_bogdanov1 Great technical question. Vector search helps with recall but doesn't solve the categorization consistency problem. You need structured vendor-to-category mappings that update deterministically, not probabilistic retrieval. We tried the RAG approach early on - it hallucinates categories at scale. The data layer we built is more like a traditional database with LLM-powered onboarding, not a vector search.
Nice launch! I run a single-member US LLC. Most of my expenses are SaaS subscriptions, a few contractor payments, and the occasional travel. Pretty clean Schedule C. Two questions: (1) How does Jupid handle the personal vs. business split on a single bank account — does it learn which recurring charges are business over time, or do I have to tag everything manually upfront? (2) For a low-volume consulting LLC (~50-100 transactions/month), is there enough signal for the categorization engine to be useful, or does this shine more at higher transaction volumes?
@sergey_kalachev Our topmost goal is eventually make it so you don't need to tag anything manually!
think of working with a real human accountant: sometimes he'll have to ask you questions ("Did you fuel your car here for personal needs or business travel?"), but often, knowing you, your company operations, your history and preferences, he can make this assumption on his own.
That's our target: Jupid should become well-versed in your operations so it'll bother you only when it's necessary and just once.
Venture Roof
The 'Works in Claude Code' angle is interesting. What does that actually look like in practice?
@dimakim You open Claude Code in your project, and you can ask things like 'what were my top 5 expenses last month?' or 'am I on track for quarterly estimated taxes?' - using your real bank data, not hypotheticals. It's your financial data available as context wherever you code. For founders who live in the terminal, it means never switching to a separate accounting app.
minimalist phone: creating folders
Just in time!! This day, we have the last day to report our incomes and calculate tax rates. Does it work internationally, or only in the US region for now?
@busmark_w_nika Yes — the Schedule C filing flow is US-specific for now. But the rest of Jupid is not limited to the US: our transaction categorization + context layer works globally, so you can use it to organize transactions, build custom reports, and work with your data in Claude Code or other reporting workflows. So the tax filing form is US-only, but the core intelligence layer is not :)
AgentQL
If our company already considers using one of the alternatives, like Fondo, how are you different from them? Any highlights/number differences?
I did all my taxes this year with Claude walking me through the whole process. It was honestly perfect. So seeing someone build a dedicated product around this makes total sense. The messy part was always the data, not the AI itself. Curious how Jupid handles bank connections for EU accounts?