DEV Community: Piotr Sarna

Backscroll launched on Product Hunt today

Piotr Sarna — Wed, 10 Jun 2026 07:24:31 +0000

We're live on Product Hunt today: Backscroll

Quick background: I kept losing useful AI conversations. I'd worked through a good explanation in Claude or ChatGPT weeks ago, and the search in each assistant only covers that one tool — no cross-history view, no semantic search.

Backscroll is the fix: export your history from ChatGPT, Claude, and Gemini, drop it in once, and search everything by keyword or meaning. It cites answers back to the original thread.

If you've got AI chat history accumulating, the demo is at https://backscroll.xyz/demo.

Would love feedback on what'd make it actually useful day-to-day.

showdev #ai #productivity #buildinpublic

I built searchable memory for my ChatGPT, Claude & Gemini history

Piotr Sarna — Tue, 09 Jun 2026 10:12:38 +0000

We have all had genuinely useful conversations with ChatGPT, Claude, or Gemini and then watched the good ones scroll into oblivion. I kept solving the same problem twice because I could not find that one thread from three weeks ago.

So I built Backscroll: export your history from each assistant, upload it once, and it becomes searchable by keyword and by meaning. You can also ask a question and get an answer pulled straight from your own past chats, cited back to the exact conversation.

How it works

Import your ChatGPT / Claude / Gemini export (the file each lets you download).
It chunks + embeds every message (Supabase pgvector, gte-small 384-dim, via a keyless edge function).
Search is hybrid: Postgres full-text keyword + vector similarity, merged.
"Ask your history" runs retrieval, then an LLM (via OpenRouter) answers with citations to the source chat.

Stack

Next.js + Tailwind, Supabase (Postgres + pgvector + Auth, RLS on every table), OpenRouter for the LLM. No background recording, no screen capture — just the exports you choose.

There is a no-signup demo so you can try the search on a sample archive first: https://backscroll.xyz/demo

Would love feedback from devs who live in AI chat tools — what do you go back and look for most? Code snippets, decisions, prompts?

I built StudyLoop: paste your notes, get AI flashcards with spaced repetition

Piotr Sarna — Tue, 09 Jun 2026 07:40:46 +0000

I kept re-reading my notes and forgetting them anyway. Active recall plus spaced repetition is what actually works, but making flashcards by hand was tedious enough that I never stuck with it.

So I built StudyLoop — you paste your notes (a chapter, lecture notes, an article) and it generates clean question/answer flashcards, then schedules them with spaced repetition so each card comes back right before you'd forget it.

The stack

Next.js (App Router) on Vercel
Neon (serverless Postgres)
An OpenAI-compatible LLM (OpenRouter / claude-haiku) for card generation
Magic-link auth (no passwords) + Stripe for Pro

How it works

You paste notes, the LLM returns a JSON array of { front, back } cards
Cards are stored with SM-2-lite spaced-repetition fields (ease, interval, due date)
Study mode surfaces due cards; you rate Again / Hard / Good / Easy and the schedule updates

It's live and free to try (3 decks, nothing to install): https://studyloop-jelonmans-projects.vercel.app

Genuinely after feedback: does "paste notes, get a deck" fit how you actually study, or is making the cards yourself part of the value? And for anyone who's launched a study tool, what got you your first users?