Penna learns your household once — who's who, what's true about each person — then fills any government, financial, or medical form for you. You just review before anything is submitted.
Starting with family immigration paperwork. No spam — one note when early access opens.
Upload a document to see your household mapped out.
Upload one or more forms you want data extracted from — PDFs, scans, or photos. Uploading several together lets Penna cross-reference names and relationships across them, so upload related documents at once when you can.
Check what Penna read from the document and fix anything that's wrong before it's added to your household.
No one extracted yet — upload a document to get started.
Upload a blank fillable PDF form. Penna reads its real form fields and fills them from your household — you review and correct every value before anything is written to the final file. Works with fillable PDF forms; a flat scan with no real form fields isn't supported yet.
Every file you've uploaded, with its review status. Files are stored privately — only you can open them.
No documents uploaded yet.
Most tools remember one person. Penna remembers your household — your spouse, your kids, who depends on whom — so every form after the first one takes minutes, not hours.
Passports, certificates, old forms — however messy. No need to organize anything first.
It builds a picture of who's who — relationships, facts, dependents — from what you gave it.
Every field gets filled from the right person's data. You check it over — Penna never submits on its own.
Visa and family-petition forms ask the same facts across dozens of pages, and getting them wrong is costly. It's where the pain is worst — and where Penna starts.
We're onboarding a small number of households before opening more broadly. Leave your email and we'll reach out.
Every family that's gone through a visa petition, a benefits application, or a mortgage knows the real cost isn't any single form — it's typing the same names, dates, and facts into different boxes, over and over, hoping this time you don't make a transcription mistake that costs you months.
Penna started from a simple observation: most tools that help with paperwork remember one person. Real households aren't one person. They're spouses, kids, dependents, and the tangle of facts that connect them — and that structure gets thrown away every time you start a new form.
So Penna remembers the household instead. Upload your documents once, and Penna builds a picture of who's who. From there, every form — starting with family immigration paperwork, the place this pain is worst — gets filled from that same understanding, with a human checking the result before anything is ever submitted.
We're early. Right now Penna is a small, focused product built around one wedge: making family visa paperwork faster and less error-prone. Everything after that — other government forms, financial paperwork, healthcare intake — builds on the same household data and the same principle: verify before anything is final.
This is a draft policy for an early-stage product and has not been reviewed by a lawyer. Replace this with counsel-reviewed language before handling real user data at any scale — this is a starting point, not legal advice.
When you create an account, we collect your email address and, if you sign up with Google, basic profile info Google provides. When you upload a document, we store the file itself and the personal data Penna extracts from it — names, dates of birth, nationality, passport numbers, and the relationships between people in your household.
Uploaded documents are sent to Anthropic's Claude API to extract structured household data. Extracted data, your documents, and any corrections you make are stored in our database (hosted on Supabase) so Penna can pre-fill future forms and improve accuracy over time. We do not sell your data or use it to train models outside of what's needed to run the product for you.
Your household data is private to your account, enforced at the database level — no other user of Penna can see it. Our infrastructure providers (Supabase for storage/database, Anthropic for document processing, Google for optional sign-in) process data on our behalf under their own security and privacy terms.
You can edit or delete any person in your household, remove uploaded documents, and correct any extracted field at any time from your dashboard. You can request full account deletion by contacting us — see the Contact page.
We keep your data for as long as your account is active. If you delete a person or document, the underlying record is removed, though a log of corrections may be retained in de-identified form to improve extraction accuracy for everyone.
Data is protected with row-level access controls so only you can read or write your own household data. Documents are stored in a private, non-public storage bucket. No system is perfectly secure, and we'll disclose any breach affecting your data as required by law.
Questions, feedback, or something not working right — we'd like to hear about it.
These are placeholder addresses — swap in your real support inbox before launch.