Snap receipts. Track spending. No bank logins. Privacy first
Recibbi turns your receipt photos into detailed spending intelligence. Upload or text your receipt photo and Recibbi reads every item, decodes register abbreviations, and adds images, sizes, categories, and unit prices. It builds your Personal Inflation Index, detects shrinkflation, compares nearby store prices, and flags overcharges. Share itemized links with your household, track trends and budgets, and export your data — all without bank logins.
34 people supporting this launch
4.8
6 ratings
Engineering VP at Recibbi Labs
AI video with native sound, paid per clip. Nothing stored.
A calmer Mac menu bar, one click away.
Enterprise Software. Intelligent Automation. Global Impact.
Comments
38Interested in developing AI agents, webapps, and AI workflow.
I like the product a lot, and I use it in my daily life. I have also purchased its 1$ plan
Engineering VP at Recibbi Labs
Thank you for your business!
good idea, useful tool~
Engineering VP at Recibbi Labs
Thank you for your kind words. Here to help!
Your product is good and a unique solution
Engineering VP at Recibbi Labs
Thank you!
The no-bank-login choice is what sells it for me — I build financial calculators with the same privacy-by-architecture principle (compute locally, collect nothing), and it's rare to see a spending tracker take that route since everyone defaults to Plaid. Question about the Personal Inflation Index: roughly how many receipts does it need before the index is statistically meaningful? Official CPI baskets smooth over thousands of prices, so I'm curious how you handle the noise of one household's shopping.
Engineering VP at Recibbi Labs
The distinct aspect of our inflation index is that it can be meaningful with your first receipt. We accomplish this by modeling the index as a combination of the general inflation index from the broader economy, aggregated and anonymized index from users with similar spending patterns, and your distinct purchases.
The no-bank-login choice is what sells it for me — I build financial calculators with the same privacy-by-architecture principle (compute locally, collect nothing), and it's rare to see a spending tracker take that route since everyone defaults to Plaid. Question about the Personal Inflation Index: roughly how many receipts does it need before the index is statistically meaningful? Official CPI baskets smooth over thousands of prices, so I'm curious how you handle the noise of one household's shopping.
Engineering VP at Recibbi Labs
The distinct aspect of our inflation index is that it can be meaningful with your first receipt. We accomplish this by modeling the index as a combination of the general inflation index from the broader economy, aggregated and anonymized index from users with similar spending patterns, and your distinct purchases.
The privacy-first approach here is the right call for financial data — this is the same instinct that keeps me away from anything that wants bank-level access just to track receipts. Question on the shrinkflation detection: are you comparing unit prices against the same product/UPC over time, or inferring it from size/weight text on the receipt itself? That seems like the harder problem once packaging redesigns happen mid-cycle.
Engineering VP at Recibbi Labs
Thank you! That's a great question Quratulain! We have a US-wide product database in the backend and match up the receipt prices from the receipt to the database prices. The database has numbers from user cohorts, store-wide averages, and national averages. It is a hard problem but we are a great team with the right skills to solve it!
Software engineer
Looks cool, will try it in action. If it actually privacy focussed and doesn't sell aggregated analytics to 3rd-parties - kudos for that, but I suspect this will be the main revenue driver eventually. Geo-tagged receipts could be appealing for marketers
Engineering VP at Recibbi Labs
Thank you!
Founder | neuralwork.net
This is a neat idea. I like that it works from receipts instead of requiring bank access. How well does it handle receipts with faded text or unusual store formats?
Engineering VP at Recibbi Labs
Thank you! For the unusual store formats and faded text we have a data processing pipeline with agentic AI based character recognition to ensure we cover different interpretations. The best scoring interpretations win. The most difficult cases also get anonymized human review. So we do our best to make sure you are happy with the result!
Founder at Stacklance, building consumer iOS apps like ScreenFine.
The "no bank logins" angle is underrated: plenty of people won't connect accounts to a budgeting app, so parsing the receipt itself sidesteps the trust barrier entirely. The Personal Inflation Index and shrinkflation detection are the standout ideas, that's item-level insight raw transaction data can't give you. My one question is OCR accuracy on faded thermal receipts and cryptic register abbreviations, that feels like the make-or-break for the whole experience. How's it holding up there?
Engineering VP at Recibbi Labs
Thanks for the thoughtful comment! Yes, we intentionally designed privacy first. The OCR accuracy gets better all the time. Did you hear how AI powered OCR, like ours, recently helped to read formerly unreadable scrolls from Pompei? If OCR can read that, it can read receipts!
i write a16z speedrun scout checks within 24hours | Co-Founder & CTO at Dial AI
Receipt friction solved with genuine intelligence. Most budgeting tools force manual categorization or rely on bank transactions that miss cash purchases. Recibbi's line-item extraction plus shrinkflation detection changes the game - suddenly you see where money actually goes and catch price creep others miss. Personal Inflation Index is especially clever, turns abstract spending into data you understand. Privacy-first approach means household spend stays yours. Real win for people who want spending awareness without surveillance.
Engineering VP at Recibbi Labs
Thank you for the comment! We are privacy-first and open-source. Those are the ways to build genuine trust with the customers.
Really useful tool, especially as someone who is trying to save money where I can.
Engineering VP at Recibbi Labs
Thank you! Spot on!
Builder of things
Nice one. I just gotta remind myself to always snap though because I be forgetting
Engineering VP at Recibbi Labs
Thank you! We are working on a reminder feature for a smartphone call for when you leave a major store like Costco.
Creator of OCRStack, an online OCR workspace for PDFs, images, and documents.
Line-item extraction plus unit-price history is much more useful than category totals alone. I like that it avoids bank logins. How does Recibbi handle faded thermal receipts and normalize the same product across different store abbreviations?
Engineering VP at Recibbi Labs
Thank you! For the unusual store formats and faded text we have a data processing pipeline with agentic AI based character recognition to ensure we cover different interpretations. The best scoring interpretations win. The most difficult cases also get anonymized human review. So we do our best to make sure you are happy with the result!
Building things to make life easier
Does this integrate with banking apps?
Engineering VP at Recibbi Labs
Thank you for the question! It is in your control. You can export data from Recibbi and import using CSV into your bank app if it allows that. We don't integrate directly.
The no-bank-login angle is the right call. Most spending trackers demand full account access just to do categorization, so doing it from receipt photos is a much saner privacy tradeoff. Line-item shrinkflation detection is genuinely useful too, since that is invisible in normal bank-feed budgeting apps. Curious how well the register-abbreviation decoding holds up on receipts from small local shops rather than big chains.
Private QR photo sharing for weddings and events
Congrats on the launch! The shrinkflation detection is a genuinely clever angle: most budgeting apps only see totals, so item-level tracking of package sizes is real added value. Two questions: how does the Personal Inflation Index handle the same product bought across different stores? And since you lead with privacy and no bank logins, are receipt photos stored after extraction or discarded?
Engineering VP at Recibbi Labs
Thank you! The index is calculated through our database that is normalized across US-wide product database in the backend and matches up the receipt prices from the receipt to the database prices. The database has numbers from the user, user cohorts, store-wide averages, and national averages. Since each line item is mapped to a product, we can track it overtime as it is purchased by many users across many stores. For privacy, the free tier automatically deletes the receipts after 24 hours. If you are on a paid plan, you have the option of storing the photo on Recibbi indefinitely if you choose to do so.
Data science
recibbi has great potential and offers some useful features, but I encountered a few issues that could be improved. Please contact me so I can share detailed feedback and suggest the changes I'd like to see. WhatsApp: https://wa.me/447307349530 Telegram: t.me/rforrank
Engineering VP at Recibbi Labs
Thank you!
Builder of things
Nice one. I just gotta remind myself to always snap though because I be forgetting
Decoding register abbreviations is the underrated hard part here. Spanish supermarket receipts print things like "P.LECH ENT 6x1L" and no generic OCR knows what that means, so if your model actually resolves those to real products with images and unit prices, that alone is worth the app. Do you handle non-English receipts yet, or is the abbreviation dictionary per country? I would happily feed it a stack of Mercadona receipts to test.
I'm here to share my experience and some tools to get into a househack!
Accountant here. A lot of people I have worked with abandon apps that want account linking, either over privacy or because the links break every other month, and receipts are the ground truth anyway. Line items also catch unit-price creep that a bank feed flattens into one number. My question is how you handle the messy half of real spending, cash purchases and the receipts nobody kept. Even a quick manual add would keep the totals complete.