<!-- mobian-agent-page publisher="dailydev" canonical="https://daily.dev/posts/a-local-ai-reads-my-kid-s-school-papers-so-i-don-t-have-to-exqaa3vky" -->

---
title: A Local AI Reads My Kid’s School Papers So I Don’t Have To
description: Backpack is a local app that reads photos of kids' school papers and turns them into a calendar and to-do list. It runs Gemma 4 through Ollama entirely on your...
canonical: https://daily.dev/posts/a-local-ai-reads-my-kid-s-school-papers-so-i-don-t-have-to-exqaa3vky
twitter:card: summary_large_image
twitter:site: @dailydotdev
og:type: website
og:site_name: daily.dev
og:title: A Local AI Reads My Kid’s School Papers So I Don’t Have To | daily.dev
og:description: Backpack is a local app that reads photos of kids' school papers and turns them into a calendar and to-do list. It runs Gemma 4 through Ollama entirely on your...
og:url: https://daily.dev/posts/a-local-ai-reads-my-kid-s-school-papers-so-i-don-t-have-to-exqaa3vky
og:image: https://api.daily.dev/og/posts/ExqAa3vKy.png
og:image:alt: A Local AI Reads My Kid’s School Papers So I Don’t Have To
og:image:width: 1200
og:image:height: 630
og:locale: en
---

> ## Documentation Index
> Fetch the complete documentation index at: https://daily.dev/llms.txt
> Use this file to discover all available pages before exploring further.

# A Local AI Reads My Kid’s School Papers So I Don’t Have To

**[Awesome Go](https://daily.dev/sources/awego)** · 8 min read · 0 upvotes · 0 comments

## Summary

Backpack is a local app that reads photos of kids' school papers and turns them into a calendar and to-do list. It runs Gemma 4 through Ollama entirely on your own laptop, so no data leaves the machine. After prompt tuning and validation checks in Go, it hit 13 of 13 dates and 8 of 10 to-dos on ten test papers, with 4 extra lines that were not on the papers. The larger gemma4:26b model found all 10 to-dos but took 3.6 seconds per paper and an 18 GB download versus 7 GB for the default e4b. The whole app is one Go binary with an embedded page, no framework, listening only on 127.0.0.1.

## Full article

daily.dev links to this article rather than hosting it. Read it at the original source: <https://dev.to/nazar-boyko/a-local-ai-reads-my-kids-school-papers-so-i-dont-have-to-2knk>

## Questions this post answers

### How accurate is the small Gemma 4 e4b model compared to the larger 26b version for structured data extraction tasks?

On a set of ten test school papers, gemma4:e4b found 13 of 13 dates and 8 of 10 to-dos after prompt tuning, while the larger gemma4:26b found all 10 to-dos but ran about 3.6 seconds per paper instead of roughly 2 seconds, and required an 18 GB download versus 7 GB for e4b.

_Comparing small versus large open-weight models for your own task gets easier with benchmarks like this on daily.dev._

### How can I run Gemma 4 locally with Ollama from the command line?

Install Ollama, then pull the model with 'ollama pull gemma4:e4b', which downloads about 7 GB once. Ollama exposes a chat API that can be constrained with a JSON schema via the 'format' parameter, so the model's response is forced into a single consistent shape, useful for building reliable local AI tools.

_Developers setting up local LLM workflows can track tool combos like Ollama and Gemma on daily.dev._

### What are some ways to keep a locally running AI app from leaking data to other hosts or websites?

Restrict the server to listen only on 127.0.0.1, reject any requests naming a different host, and reject cross-site requests from other websites open in the same browser. A test in the codebase can fail the build automatically if the code ever references a host name other than the local machine.

_Anyone hardening a local-first app's network boundary can follow patterns like this on daily.dev._

## Similar posts on daily.dev

- [I built a private AI study assistant that runs entirely on my laptop](https://daily.dev/posts/i-built-a-private-ai-study-assistant-that-runs-entirely-on-my-laptop-rkruwx0ax) · XDA Developers · 0 upvotes · 0 comments
- [Availazy: Gemma 4 reads the volunteer group chat on my laptop, Python does the dates, and I just say yes or no](https://daily.dev/posts/availazy-gemma-4-reads-the-volunteer-group-chat-on-my-laptop-python-does-the-dates-and-i-just-say-btbyef93h) · DEV · 1 upvotes · 0 comments
- [This open-source Google app turned my phone into a fully local AI lab](https://daily.dev/posts/this-open-source-google-app-turned-my-phone-into-a-fully-local-ai-lab-d9txzwke5) · XDA Developers · 2 upvotes · 0 comments

---

Tags: [#golang](https://daily.dev/tags/golang), [#privacy](https://daily.dev/tags/privacy), [#local-ai](https://daily.dev/tags/local-ai), [#ollama](https://daily.dev/tags/ollama), [#gemma](https://daily.dev/tags/gemma)

[View this post on daily.dev](https://daily.dev/posts/a-local-ai-reads-my-kid-s-school-papers-so-i-don-t-have-to-exqaa3vky)

```json
{"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]}
{"@context":"https://schema.org","@type":"TechArticle","headline":"A Local AI Reads My Kid’s School Papers So I Don’t Have To","url":"https://daily.dev/posts/a-local-ai-reads-my-kid-s-school-papers-so-i-don-t-have-to-exqaa3vky","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/a-local-ai-reads-my-kid-s-school-papers-so-i-don-t-have-to-exqaa3vky"},"datePublished":"2026-10-30T04:45:57.923Z","dateModified":"2026-10-30T04:46:28.510Z","description":"Backpack is a local app that reads photos of kids' school papers and turns them into a calendar and to-do list. It runs Gemma 4 through Ollama entirely on your...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/853b6123ab40a8b508c499b13f02665b?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/853b6123ab40a8b508c499b13f02665b?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Awesome Go","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Awesome Go","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/awego","url":"https://daily.dev/sources/awego"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/a-local-ai-reads-my-kid-s-school-papers-so-i-don-t-have-to-exqaa3vky","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"golang,privacy,local-ai,ollama,gemma","timeRequired":"PT8M"}
{"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Awesome Go","item":"https://daily.dev/sources/awego"},{"@type":"ListItem","position":3,"name":"A Local AI Reads My Kid’s School Papers So I Don’t Have To"}]}
{"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/a-local-ai-reads-my-kid-s-school-papers-so-i-don-t-have-to-exqaa3vky#faq","mainEntity":[{"@type":"Question","name":"How accurate is the small Gemma 4 e4b model compared to the larger 26b version for structured data extraction tasks?","acceptedAnswer":{"@type":"Answer","text":"On a set of ten test school papers, gemma4:e4b found 13 of 13 dates and 8 of 10 to-dos after prompt tuning, while the larger gemma4:26b found all 10 to-dos but ran about 3.6 seconds per paper instead of roughly 2 seconds, and required an 18 GB download versus 7 GB for e4b. Comparing small versus large open-weight models for your own task gets easier with benchmarks like this on daily.dev."}},{"@type":"Question","name":"How can I run Gemma 4 locally with Ollama from the command line?","acceptedAnswer":{"@type":"Answer","text":"Install Ollama, then pull the model with 'ollama pull gemma4:e4b', which downloads about 7 GB once. Ollama exposes a chat API that can be constrained with a JSON schema via the 'format' parameter, so the model's response is forced into a single consistent shape, useful for building reliable local AI tools. Developers setting up local LLM workflows can track tool combos like Ollama and Gemma on daily.dev."}},{"@type":"Question","name":"What are some ways to keep a locally running AI app from leaking data to other hosts or websites?","acceptedAnswer":{"@type":"Answer","text":"Restrict the server to listen only on 127.0.0.1, reject any requests naming a different host, and reject cross-site requests from other websites open in the same browser. A test in the codebase can fail the build automatically if the code ever references a host name other than the local machine. Anyone hardening a local-first app's network boundary can follow patterns like this on daily.dev."}}]}
```